|Ask the Experts|
February 21, 2018
Pros and Cons of Cleaning No-clean
We are working with our contract manufacturer to develop the best process for our boards. With today's newer chemistries, what are the pros and cons of cleaning circuit boards assembled using a no-clean process?
|Expert Panel Responses|
This is a great question, and one that too few engineers think about. After all, there are lots of smart guys at solder companies making some pretty strong claims about the performance of their fluxes, and touting the cost-savings derived from skipping the cleaning process. So, why WOULD anybody want to clean no-cleans? Well, it happens all the time. Probably half of the PCB assembly companies across the globe are using "no clean" fluxes and still cleaning. There's lots of good reasons why. Here's a few:
Points to consider:
Application Technology Manager
The pros & cons of cleaning circuit boards are monumental in opinions and applied science from both sides of the camp fire. We are now in the era of "no-clean". However, this term seems to be an oxymoron as it relates to an expectation that soldering is the only operation in the electronic assembly process. It is unrealistic to expect electronic assemblies to remain absolutely clean throughout the entire manufacturing life-cycle. Subsequent operations like inspection, touchup, repair/rework, testing, intermediate handling, storage, and final assembly are performed inclusive to soldering operations. Most of these operations are not carried out under clean room conditions by people who always wear gloves or finger cots. The assemblies thus become contaminated by human handling, sedimentation from the air, natural oxidation, and contact with workbenches, storage bins, and so on. The level of cleanliness after no-clean soldering bears little resemblance to that existing at the final stage of assembly. Will this be acceptable? Is there value added to clean or not clean? Are your assemblies for garage door openers with a $4 dollar BOM or a board as part of an assembly costing hundreds or thousands of dollars? I know what I will be cleaning! Ineffective cleaning has been identified, using several reliability test procedures, as a major cause for component and circuit failures. In the era of no-clean, violations of the rules of cleanliness were not manifested as major failures and the industry grew complacent. There seems to be a basic lack of appreciation as to why we clean. This no-clean oxymoron has left the electronics industry with a legacy of declining product yields and an increase in field failures. We must remember that in the electronics industry, the basic technical reasons for cleaning are as follows:
Technical Expert Sales Support
* No-clean pastes have been used for the last 20+ years and have had a very good track record
* 70-75% of global electronics is a no-clean process with no cleaning of the flux residue - big $$$ saving
Cons of cleaning no-clean flux residues
* DI water alone cannot be used to clean & it will react with the rosin (in the flux residue) to form white powder; solvent-saponifier chemistry needs to be used and this means more $$$; also capital expenditure is involved if there is no cleaner in the first place
* for low-standoff components, it becomes more difficult to clean
* the rosin in the no-clean flux provides the inert, non-hygroscopic cover on top of reflowed flux residue and ensures that the ions do not have a path to electromigrate and cause corrosion. If the no-clean flux residue is not cleaned off properly, then it is as bad as leaving water-wash flux residue behind. Ions left behind on the board will cause corrosion
Potential Pros of cleaning no-clean flux residues
* for post-reflow operations such as underfilling / conformal coating, the no-clean flux residue id left behind may outgass / prevent good adherence with the epoxies and therefore to the substrate; though there are many many applications that underfill / conformal coat over the no-clean flux residue (without cleaning) with no issues
*allows for good cosmetics though a lot of no-clean flux residues are almost colorless; also cosmetics is subjective and en-reliability is most important.
Technical Manager - Europe
There are lots of reasons to remove no-clean flux residues from PCBs. Cleaning came to no-clean fluxes when problems emerged with in-circuit testing. The residues from the first no-clean formulations were gummy, and test pins could not reliably penetrate the flux to make electrical contact with the test pads. In addition, the residue would gradually build up on the pins, causing problems test accuracy and the increase of maintenance was also a draw back. However, the most of the newer formulations of no-clean solder paste have negated this issue. Today, one of the most common reasons to remove no-clean fluxes is to prevent malfunctions in circuits with clock speeds over 1 gigahertz. "When you get above 1 gigahertz, the electronscan conducted on the outer surface of the conductor. If you have flux on those conductors that have those high clock speeds, it can interfere with these fast signals." Another reason to remove no-clean fluxes is to increase the bonding of conformal coatings and under fill materials. Some no-clean flux residue can absorb moisture so when you dispense an under fill material beneath a flip chip, and heat it to curing temperature, you can get a little pocket of steam or gas. Eventually, that can cause the under fill material to separate from the board. Some very conscientious houses will clean no-clean fluxes to remove any possibility of solder balls. They will also clean PCBs exposed to extreme environments; or simply to improve the cosmetic appearance of the boards.
Electronic Technology Corporation
Cleaning a no-clean flux requires both, time and effort. It is necessary to match the chemistry to the flux, all of which is added cost. In terms of reliability, I have no data comparing cleaned with uncleaned, however partially cleaned flux residues can release ionic material that can compromise reliability. Cleaning is most advantageous when an additional material is to be applied to the assembly (e.g. conformal coating, potting material, etc.). However, the removal of residues eliminates the possibility of adverse interactions between the flux residues and the secondary material.
I believe that there are a lot of new cleaning chemistries in the market today that do clean off no-clean residues. Kyzen, who makes cleaning solutions would be a good company to talk with. Typically No Clean residues are difficult to remove since they aren't very soluble in most solvents including water. I would typically suggest that if you decide to go down the route of no clean technologies you try to not clean.
In my opinion, "no clean" is a myth, at least the cases I have encountered. There are still oxidation residues (albeit trace quantities) and by products caused during the heating cycles. In my view, the more critical the application is (in terms of contamination potential from ANY source), cleaning post-process is still the wise path. I guess the balancing act is what is the cost of product failure, for whatever reason; and, what is the likelihood of contamination from a "no-clean" process. Again, I think that "no-clean" is a marketing over-hype.
If it is "no-clean", why do you clean it?Cleaning the "no-clean" residues is not only costly (chemicals $$$) but riskytoo. With the low profile components, high complexity boards and assembliesthat need conformal coating, I am a big fan of water soluble materials. From my perspective, depending on the manufacturingprocess, you can still use the clean or the no clean solder. My recommendationis to evaluate all the potential risks before you start the process. Fluxresidues usually create the opportunity for late failures.
Engineering and Operations Management
I would like to change the question from "Pros and Cons of Cleaning No-Clean" to "Pros and Cons of Cleaning".
It is true that all no-clean fluxes and pastes leave behind some residues. In some cases, the residues are tolerable to the assemblies and in other cases they are not. In a recent industry pole of assemblers reflowing with no-clean paste, 53% of respondents indicated they clean the assemblies after reflow. Perhaps we should consider a more accurate description of no-clean flux/paste as "Low Residue".
The bigger issue when an assembler reflows an assembly and does not remove the flux residues is the fact that nothing else is removed. One should consider that a cleaning process cleans everything. There are contaminant species from board fabrication, component fabrication residues, assembly residues, human handling residues, and, of course, flux residues. When we choose not to remove flux residues, we choose not to remove any residues. A defluxing process is more accurately defined as a cleaning process.
Every electronic assembly has a tolerance for residues, some have a higher tolerance and some have a low tolerance.
Assembly design, component placement and density, in-use environment, voltage, etc all play a role in determining an assembly's residue tolerance. The result of intolerable residues is electro-chemical migration (ECM). If one wants to eliminate the possibility of ECM, remove all residues (flux and others).
The feedback is one-sided. I would like to hear from the Flux manufacturers who must have data showing no increase in failures by using No-Clean. 53% of board stuffers in a recent Poll (highlighted in todays newsletter) said they clean no clean that still means 47% do not. If there were such danger and defect levels with not cleaning I would expect that the percentage of Board cleaners after no clean fluxes would be much higher. Please show the data that tells us we have to clean? is it a 1% difference or a 40% difference in the defects?Dave Chapman, Coilcraft
|Submit A Comment|
Free Newsletter Subscription
Circuitnet is built for professionals who bear the responsibility of looking ahead, imagining the future, and preparing for it.
Insert Your Email Address