By Dan Kay

Brazers commonly encounter voids in brazed joints and often wonder where they come from and how to avoid them in future brazements.

Some common sources of voids in braze joints are:

1. Surface contamination
2. Base metal and brazing filler-metal (BFM) constituents
3. Brazing methods/temperatures used
4. Poor joint fitup

The first three items listed above can often result in gas bubbles being formed in brazed joints. Such gas-bubble voids will usually try to form in spherical shape as they move through a joint. The "rounded" edges of such bubble-voids can often be clearly seen in cross-section photomicrographs of brazed joints, especially under high magnification. The inside surfaces of a bubble-void will often appear "clean" or "shiny" as well.

Let's briefly look a little more closely at these sources of voids in brazed joints with a look at surface contamination.

1. Surface Contamination

It is not uncommon to hear some brazers say, "Don't worry about surface contamination – the furnace will take care of it," or "Don't worry about surface contamination – just put more flux on the part to take care of it." Both statements are dangerous and can lead to weakened joints and, in many cases, to failed joints.

Oils, greases, lubricants, dirt, etc. left on the surface of assembled parts in a brazement often contain ingredients that will tend to volatilize and outgas at the elevated temperatures involved in brazing. These "gases" will attempt to expand and move out of the confining braze joint, and, as long as there is an "escape path" open to the outside atmosphere for these expanding gases, they will not tend to be a problem during brazing. Since contaminants on the outside surface of parts being brazed are readily open to the atmosphere, such contaminants may completely volatilize, causing no problems to the brazement.

When the inside surfaces (faying surfaces) of a brazement are contaminated and not cleaned off prior to assembly, however, it frequently becomes impossible for those contaminants to volatilize (turn to a gas). Even if much of it could volatilize, those gases usually find it difficult, if not impossible, to make way to the outside of the joint where they can finally be released to the atmosphere. They will become entrapped (as bubbles) inside the brazed joint when the joint finally solidifies.

Additionally, be aware that surface contaminants and oxides inside a joint that do not volatilize can ruin a braze! Since molten brazing filler metal (BFM) does not want to bond to (or flow over) oils, dirt, greases or oxides on the faying surfaces inside a joint, the BFM may not be able to penetrate through the braze joint at all when surfaces have not been adequately cleaned prior to joint assembly.

Thus, it is very important that all surfaces to be brazed must be cleaned PRIOR to assembly for braze. Surfaces must then be handled with gloved hands (so that they are not recontaminated by the brazer's fingers and hands). These precautions will minimize any voids in the brazed joint.

Let's briefly look a little more closely at the second source of these voids in brazed joints, base metal and brazing filler-metal constituents.

2. Base Metal and BFM Constituents

Many base metals and BFMs have constituents in them that can easily volatilize when heated to brazing temperatures. Zinc, cadmium and lead are three such metals that will, in fact, outgas readily during any kind of brazing process, and proper precautions should be observed when such metals are used in brazing.

Lead may be found in some steels or brasses to enhance the machinability of those base metals. If such metals are then used as part of a brazement, however, the lead will quickly outgas, forming bubbles in the brazed joint and perhaps leaving holes in the base metal (such holes might result in leak paths through the metal, hurting hermeticity of the brazed assembly). This can obviously become a problem where leak-tight hermetic-seals are required for brazements in service.

Zinc and cadmium are used in certain BFMs as temperature depressants to help lower brazing temperatures and also to enhance the flowability of these BFMs on certain base metals. Zinc and cadmium may also be found as platings on some metal parts that are to be brazed. Zinc and cadmium, like lead, will readily outgas upon heating to brazing temperatures, and this will be seen as bubbles in the braze joint or as fumes in the workzone of the brazers. Neither situation is desirable.

Please bear in mind that because such outgassing of these three metals will, in fact, occur during heating to braze temps, they should NEVER be used in any vacuum brazing environment as they will contaminate the vacuum furnace and its pumping system, perhaps to the point of rendering the furnace non-usable! I strongly recommend that metals and BFMs containing these three elements be limited to flame brazing and induction brazing and that they not be used in any kind of furnace brazements because of the potential for contamination of furnace surfaces. Even with flame (torch) brazing or induction brazing, proper venting of the brazer's breathing zone must be done so that the brazer does not have to breathe the fumes being generated.

Let's briefly look a little more closely at the third source of these voids in brazed joints, brazing methods/temperatures used.

3. Brazing Methods/Temperatures Used

It is very important to control the brazing temperature and time as much as possible to minimize outgassing of metal constituents that can form bubbles in a brazed joint. It is not difficult to control the temperatures of furnace brazing since it can be programmed to within a few degrees of desired temperature. It is usually more difficult to control actual temperatures involved in flame brazing than for furnace brazing.

When brazing temperatures are allowed to go too high, there is a strong thermodynamic driving force to move the metal from solid to liquid to gas. I've frequently witnessed flame-brazing operations where the flame setting used by the brazer is too intense, resulting in overheating of the joint and outgassing of BFM constituents as he/she tries to "speed up the process" to get more production done. Although that person may be brazing more parts per hour, etc., the quality of such joints is open to question! Gases formed from such overheating can result in excess fumes in the breathing zone as well as lots of gas bubbles (voids) in a brazed joint.

Proper training and practice is essential to be able to bring the brazing temperature to a point where the BFM will melt and flow throughout the brazed joint but not be so high as to cause the liquid BFM to turn to a gas, resulting in imperfect joints.

Summary

To minimize the outgassing of these metallic elements, do not aggressively heat the base metals or BFMs involved but try to use temperatures that just melt and flow the BFM but no more. Do not overly extend the time of brazing either. "Get in, braze and get out." When using base metals and/or BFMs containing these elements, you will always outgas them to some extent. Your job is to minimize the outgassing as much as possible and to be sure you only use approved brazing processes for such materials (never vacuum).

Let's now look a bit more closely at the fourth source of these voids in brazed joints – poor joint fit ups – and then add a brief review and summary to this topic.

4. Avoiding/suppressing voids in brazed joints:

1. Be sure all surfaces of parts to be brazed are very clean and free from any lubricants, oils, greases or oxides that might outgas during heating.

2. When flame brazing or induction brazing, do not overheat the brazed joint or the BFM. Practice will enable the brazer to uniformly melt and flow the BFM throughout the brazed joint with a minimum of gas bubbles.

3. When furnace brazing:

a) Atmosphere continuous belt furnace – Carefully control furnace temperatures and belt speeds so that the assemblies do not overheat during their passage through the furnace. This can be determined by placing thermocouples on a set of parts going through the furnace and then examining cross-section photomicrographs of brazed joints. Modify belt speed and furnace set temperatures until cross sections show complete BFM flow with a minimum of gas bubbles in the joint.

b) Vacuum furnaces – Monitor as shown above using thermocouples, but it is also very important that the level of vacuum is such that it will not cause metallic elements to vaporize. As pressure levels get less and less in a vacuum furnace, the temperatures at which metals vaporize gets lower and lower. Standard "vapor-pressure charts" show this. It may be necessary to braze in a partial pressure of inert gas in the vacuum furnace, achieved by backfilling the vacuum furnace during the brazing process. This process, performed by many brazing companies, involves backfilling a vacuum furnace to a pressure of approximately 100 microns or more, using argon or nitrogen in order to "suppress" the outgassing of any metallic elements in the base metals or BFMs involved.

Concluding Thoughts

When examining voids in a braze joint via cross-sectional micrographs:

1. Gas-bubble voids usually tend to have rounded edges, and the inside of the voids typically appear clear and shiny.

2. Voids resulting from surface contamination can have very differing shapes and edges. The inside surfaces of the voids can be discolored and may show presence of residues or surface irregularities as compared to properly brazed surfaces. Microprobe analysis of the inside surfaces of a void can sometime pinpoint the elements present in the void to assist in determining its cause.

3. Voids resulting from poor gap clearances usually show measurable differences in the distance between the faying surfaces as compared to that of properly brazed joints in the same assembly.

4. To prevent gas-bubble voids, be sure parts are kept very clean prior to and during brazing, and be sure temperatures used for brazing are not excessive. In vacuum-furnace brazing, this may also necessitate the use of a backfilling gas to build up a partial pressure in the vacuum furnace.

To View a listing of all Dan’s articles please click here