Fig. 1 Terms used in vacuum-brazing

There are many vacuum-brazing shops out there that still believe it is necessary to try to use the strongest vacuum possible for brazing if they expect to get good results.

Such thinking is erroneous, and has led many shops to actually see “worse” results (increased void content of joints, increased discoloration on furnace walls, etc.) than they would have seen if they had merely used a “weaker” (less strong) vacuum.

Many people today still like to use some of the older vacuum terminology, such as “soft vacuum”, “rough vacuum”, “hard vacuum”, etc., as shown in Figure 1, and some of those same people still believe that a “very hard vacuum” is always necessary for effective brazing.

Important Note #1: Good Brazing Does Not Necessarily Require a Very Hard Vacuum!

How “hard” a vacuum is necessary for good brazing? Just “hard” enough to reduce the amount of oxygen present in the chamber to the level that the number of oxygen atoms remaining in the hot-zone of the furnace is not sufficient to cause damaging surface oxidation on the faying surfaces of the metals being brazed. Thus, different metals will require different levels of vacuum to achieve this, since each metal reacts with oxygen differently, some more readily than others. For instance, titanium and aluminum both react extremely rapidly with any oxygen present in the furnace atmosphere, and hold onto those oxides tenaciously. To effectively braze such base metals, one of the requirements will be to braze in a fairly hard vacuum, one in which the number of oxygen atoms is extremely small. To achieve this, it also means the necessity of being sure that the furnace is very clean, and very leak-tight!

It is not uncommon for people to hear me say in my brazing-seminars: “Braze with the weakest vacuum you can get away with!”, meaning that it is only necessary to reduce the pressure in their vacuum furnace to a level where there isn’t enough oxygen present in the vacuum atmosphere to prevent wetting (by the molten brazing filler metal) of the particular metal surfaces they are trying to braze. As noted in earlier articles, brazing filler metals (BFMs) do not like to bond to, or flow over, oxidized surfaces. Oxides can very effectively prevent any brazing from occuring.

Please understand that when you purchase a vacuum-furnace with a diffusion-pump attached to it, that vacuum furnace, when operating, will seek to achieve a level of vacuum dictated by the size of the diffusion pump. Thus, a vacuum furnace with a small diffusion pump might only be able to achieve a vacuum level of 10-4 Torr, whereas a furnace with a much larger diffusion pump (or more than one diffusion pump) might be able to pump down all the way to 10-9 Torr. The level of vacuum very much depends on the size of the diffusion pump installed on the furnace.

So, if you have purchased a vacuum furnace that has a fairly large diffusion pump attached to it, and you didn’t want to pull a hard vacuum (as determined by the large diffusion pump) you might opt to “disconnect” that diffusion pump, thus limiting your vacuum to approximately 10-3 Torr — the level achievable by operating only the mechanical pumps (the blower and rotary pump). Or, you might opt to still use the existing diffusion pump, but deliberately “leak” an atmosphere gas into the vacuum system (from argon or nitrogen bottle hoses connected to the furnace), doing so at a rate fast enough to “swamp the pumps”, i.e., faster than the pumps can remove the gas from the furnace, thereby causing the vacuum inside the furnace to weaken, and a partial-pressure of the “leaked-in” atmosphere gas to begin to build-up inside the furnace.

Vapor Pressure Curves for Metals

Fig. 2 Vapor Pressure Curves for Metals

Important Note #2: The Gaseous Atmosphere Leaked into the Furnace Must Be Very Dry, with a Dewpoint of -60°f/-50°c or Drier (as Measured Right at the Furnace)!

Just how much atmosphere needs to be leaked into the furnace, i.e., how much of a “partial-pressure” is needed?

Many brazing shops indicate that they build up partial-pressures to about 200-microns or more for their partial-pressure brazing. Why did they select such a number, and what does that actually mean?

Well, to understand that, we’ll need to look again at another chart that I’ve referred to in past articles, namely, the Vapor Pressure Curves for Metals shown in Figure 2.

As you can see from the chart, the vertical axes are labeled differently on the left side and the right side. The left-side shows pressure in millimeters of Mercury (Hg), whereas the right side shows pressure in microns. What is the significance of such pressure measurements, where did they originate, and how do they relate to the chart shown in Fig. 1? To answer this, we need to take a brief look at the primary historical event in discovering the meaning and value of barometric pressure measurement.

As illustrated in Fig. 3, Evangelista Torricelli, back in the year 1643, inverted a tube of Mercury (Hg) into a dish, and the atmospheric pressure on the surface of the Hg that drained out into the dish, in conjunction with the Torricellian vacuum created in the closed space at the top of the column, kept the Hg suspended in the tube at a height of 760 mm (29.92 inches). This 760 mm of Hg came to represent “1 standard atmosphere of pressure” (when measured at sea-level and about 25°C).

Later on, scientists decided to give recognition to Torricelli’s work by naming each one of those 760 millimeter increments a “Torr” in his honor. Therefore:

1 Torr = I mm Hg

As you can then see, one standard atmosphere of pressure can also be called a pressure of 760-Torr.

Okay, let’s now go back to the chart shown in Fig. 2. The vertical axis on the right side of the chart is measured in “microns” of pressure. 1-micron (also called 1-millitorr) is 1/1000th of a Torr. Temperature on the chart is shown
along the horizontal axes.

Notice that each curve in Fig. 2 represents a specific metal. The curve for Chromium (Cr) runs approximately down the center of the chart from upper-right down to lower-left. When operating on the left side of the curve for any given
metal, that metal will not be volatilized. However, when operating on the right side of any given curve, that metal can volatilize (out-gas). It is always desirable to operate on the LEFT side of any given curve.

Let’s take Chromium (Cr) for example. You can see on the chart in Fig. 2 that the line for Cr crosses the dotted horizontal pressure line near the top of the chart (which represents one full atmosphere of pressure) at about 4350°F/2400°C. But, as a vacuum is pulled in a furnace, and the pressure drops, the Cr line begins to move down to the left. Thus, at 10-3 Torr pressure on the left vertical axis (equals 1 micron, or 1 millitorr on the right vertical axis) the Cr line crosses the 2100°F/1150°C temperature line. At 10-4 Torr, the Cr line is at only about 1900°F/1040°C. Remember, if your temp/vac-level combination puts you on the right side of a given metal’s curve, that metal will begin to volatilize (outgas).

Torricelli’s experiment with Hg in an inverted tube

Fig. 3 Torricelli’s experiment with Hg in an inverted tube

So, if, for example, I were brazing 304-stainless steel at 2200°F/1200°C, and a vacuum level of 10-5 Torr, you can see that you would be operating far to the right side of the Cr line (304-stainless contains about 18% Cr in its chemistry), and chromium would volatilize (outgas) from the stainless steel during brazing, perhaps resulting in increased voids in the brazed joint, as well as the volatilized chromium condensing on the furnace walls, causing a blue/green discoloration.

The harder the vacuum I pull in the vacuum furnace, the lower and lower will be the temperature at which chromium (and all other metals) volatilizes. Thus it gets easier and easier to volatilize metals as you pull a harder and harder

But, I don’t want to volatilize the various constituents of the base metals or the brazing filler metal (BFM) that I am using in the vacuum furnace. So, I have to “weaken” the vacuum, making it weaker and weaker, until I reach a level of vacuum at which the particular metallic ingredient of the alloy will not volatilize! This is achieved, as I mentioned earlier, by back-filling the furnace with a dry (oxygen-free) atmospheric gas until a partial-pressure is reached at which the metallic element (chromium in our example) will not volatilize (outgas).

Back to our example with chromium. To determine the amount of pressure of back-filling gas you need to prevent chromium from outgassing, determine where the Cr-line crosses your selected brazing-temp. Thus, if you’re brazing at 2200°F/1200°C, make a mark on the chart where the Cr-curve crosses that temperature line. Now, from that mark you placed on the chart, move horizontally to the right until you cross the “microns” axis. Note that number, which is about 4-microns of pressure.

As you can see, it doesn’t require much back-filling pressure to suppress the outgassing of chromium. But, because it is not easy to control the in-flow of a gas into your vacuum furnace to achieve such a marginal amount of pressure, most brazing shops opt to increase the inflow of a dry gas (typically argon or nitrogen) to a controllable partial-pressure level of about 100-200 microns. It is generally not necessary to exceed such a pressure to completely suppress the outgassing of chromium. Looking again at the chart in Fig. 2, let’s say that you back-filled your vacuum furnace to a partial-pressure of 100-microns in order to suppress the outgassing of chromium. Are you actually well-protected with such a partial-pressure? Well, let’s see. At what temperature does that 100-micron horizontal line intersect the Cr-curve? You will notice that it occurs at about 2600°F/1430°C. Thus, with a partial pressure of only about 100-microns, it would require you to braze at about 2600°F/1430°C or higher before any chromium would start to volatilize (outgas). So, as you can see, you are indeed well-protected against any outgassing of chromium from the stainless steel.

You need to do this evaluation for each of the major metallic-elements in both the base-metal being brazed, as well as the BFM you are using. It is a very important factor that you must always consider for every vacuum-brazing run you conduct.


Braze with the weakest vacuum level you can get away with. By so doing, you will prevent the outgassing of some of the metallic components of your base-metals and BFMs, the outgassing of which can negatively affect your furnace and the quality of your brazement.

Problem solve and improve with our technical experts.