Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
Cleanliness test

Cleanliness test

Subscribe to Thread

Cleanliness test � �Area Grid Arrays�. Evaluation of resi... - Jan 02, 2003 by

Richard, With respect to the Resistivity of Solvent Extra... - Jan 02, 2003 by

Good day and thanks for your comments Mike, What did I un... - Jan 03, 2003 by

Hi Richard My experience with R.O.S.E. testers has been p... - Jan 03, 2003 by

Richard

#22889

Cleanliness test | 2 January, 2003

Cleanliness test � �Area Grid Arrays�.

Evaluation of residues� resistivity in a specific location on the board. (As compared to �Solvent Extract� evaluation.)

We are: SMT assembly, using standard �water soluble� process with micro BGAs (example: Configuration: 6x8-2, pitch: .75mm (0.029�), size: 7.28 x 6.96mm (0.287� x 0.274�), ball size: 0.35mm (0.014�). Here is my dilemma�

I am familiar with 2 methods of PCB assembly cleanliness evaluation. � Ion Chromatography, (per IPC-TM-650, method 2.3.28) � Solvent Extraction, �ROSE� test (Resistivity of Solvent Extract) (�Zero Ion� or �Omega Meter�) (per IPC-TM-650, method 2.3.25)

With both of these methods the entire specimen (panel, board) in immersed (washed) in �extracting solvent�, only an average board contamination (not area specific) is obtained.

Both methods assume a good access of the �extracting solvent� to the entire board surface. (Ion Chromatography test process attempts more aggressively to gain that access.)

Two questions arise�

What if, in the case of �BGA (or some other � laying very close to the board � CSP (Chip Scale Package)) where access for the liquid is very restricted, the actual washing � part of our manufacturing process � might be limited?

So, in these areas, where the wash quality is problematic, the evaluation of that wash is also problematic.

We can go around this by condoning to destruction some of the �test boards�, by prying (forcibly lifting, creating a gap) suspected components from the board. This breaks the board and / or the component, but opens the access (for the �extracting solvent�) to the restricted area.

The second question is� Since these �BGAs (or other � laying very close to the board � CSPs) are rather small, the quantity of (possibly) entrapped residues under them is also small. When averaged over the whole area of the board (both of the existing methods do that), the results would be quite (probably falsely) acceptable.

Entrapped residues from soldering operation are very corrosives when left on the board. To evaluate if they are there, we will lift components from the boards, (create access for evaluation), but we also need a method to test for residues in a specific spot, not just an average over the entire board�s surface.

Any comments about my concerns?

(I have looked for residues (with microscope 30x, after removing the BGA), visually found none. But I am looking for something more objective and precise)

Thanks for your time and (I hope) comments.

Richard

This message was posted via the Electronics Forum @

reply »

Mike Konrad

#22891

Cleanliness test | 2 January, 2003

Richard,

With respect to the Resistivity of Solvent Extract (ROSE) test, the test solution is exceptionally aggressive. With the Zero-Ion for example, the test solution�s resistivity is 150 M-Ohms. Additionally, because test solution is made up of a 75% IPA, 25% DI water mixture, the surface tension is far less than the rinse water used to rinse the boards after washing and therefore much more capable of penetrating under a component.

Your observations are correct with respect to the inability of a ROSE tester to indicate the precise location of contamination. If a ROSE tester indicates no or little contamination, you should rest assured that the board is clean. If a ROSE tester indicates a moderate level of contamination, then additional testing may be required to determine if the contamination was evenly distributed across the board of concentrated under a component.

Because destructive testing is expensive and may actually introduce contamination, I would not recommend it as a matter of policy. I would recommend, however, testing two sets of boards in a ROSE tester, one set with components in tact, the other with low-profile components slightly lifted then compare the results. Both sets of boards should be washed in the same batch. You should be able to establish the tester�s ability to detect contamination equally between the boards with components in tact and components lifted.

There are differences in ROSE testers. Some units direct the test solution through DI media (dynamic method) during the test to maintain solvency while others do not (static method). Some testers utilize 150 M-Ohm test solution while others use 60 M-Ohm solution. Some units utilize spray-under-immersion to agitate the boards and release trapped contamination while others simply immerse the boards into a still bath.

Mike konrad@aqueoustech.com

This message was posted via the Electronics Forum @

reply »

richard

#22900

Cleanliness test | 3 January, 2003

Good day and thanks for your comments Mike,

What did I understood from your notes�

1) ROSE test (�extracting solution�) is probably good enough (with good equipment) to penetrate the space under my micro BGA. 2) I should test 2 parallel batches of boards; one with lifted components (lifted after cleaning, of course). No issue with solvent penetration. 3) If the results are the same�all is well. (Assume, no or little contamination). 4) If the results are not the same (assuming more contamination detected on �lifted components� boards), then I am in trouble (double trouble) a. I am not washing well, and b. Can�t detect it (can, only when components are lifted)

Yet, I am still back to my original concern(s) - I am suspecting that, since micro BGAs are small, any contamination (however, dangerous when left in that small area) will get diluted to the point of invisibility. - To quote your statement: �If a ROSE tester indicates a moderate level of contamination, then additional testing may be required to determine if the contamination was evenly distributed across the board or concentrated under a component�. Any suggestions on how can I do that (evaluation of residues� resistivity in a specific location on the board)?

Richard

This message was posted via the Electronics Forum @

reply »

Mike Konrad

#22904

Cleanliness test | 3 January, 2003

Hi Richard

My experience with R.O.S.E. testers has been positive with regard to detecting ionic contamination (even exceptionally low levels). As previously mentioned, a good R.O.S.E tester will be able to detect minuet levels of ionic contamination entrapped under the smallest component. It is very important, however, the have the correct ratio of board area to volume of test solution. A very small board in a very large volume of test solution may result in a premature test termination. This can be resolved by ensuring an adequate ratio of test solution with sample area and ensuring an adequate test time (specific recommendations are brand dependant).

R.O.S.E. testers are popular because of their accuracy and the very low cost of conducting a test. If, however the R.O.S.E. tester indicates a moderate or greater level of contamination, additional testing may be required for reasons outlined previously. There are several �residue experts� available to determine (for a fee) the type and location of residues on a board. Two companies I would recommend are:

Contamination Studies Laboratories http://www.residues.com

American Competitiveness Institute http://www.aciusa.org

Mike

This message was posted via the Electronics Forum @

reply »