In this case, at isolation, it is impossible to tell the difference between a real leak and a virtual leak. At isolation, a straight leak line appears curved due to the logarithmic scale.

Vacuum leaks have led more than one technician deep down the rabbit hole, wasting hours upon hours, pressurizing the system with nitrogen—or nitrogen and a trace gas—leak testing with electronic leak detectors, and soap bubbles, removing panels to check evaporators, condensers, and other components that are buried in places that are not easy to get to nor fun to go. Hours have been spent watching a rock steady standing pressure test, leaving the technician to dream about a magical particle that is blocking some torturous leak while under pressure, and then when under a vacuum air rushes in, never letting the system reach or hold an acceptable level of vacuum. Even worse, are when those leaks win and the technician hastily dumps the refrigerant charge into an improperly evacuated system. A guarantee that someone will return for a much more time consuming and costly repair in the near future.

The evacuation can stall or rise during a decay test as refrigerant outgasses, moisture boils off, components outgas, or trapped gasses are released from o-rings, threads or flange gaskets. This outgassing can mimic an apparent or “virtual” leak in the system, something no one wants to see.

With typical vacuum gauges, there is no way to definitively tell the difference between a real and virtual leak, and there is a very good reason why. Most vacuum gauges, whether analog or digital are logarithmic scaled, meaning they have many different scales or characteristics as they move through different stages of evacuation. Even those that advertise full range single micron resolution are subject to varying accuracy, making them non-linear in nature as they approach their ideal range of measure. What this means is on a graph or LED display, the manufacturer is trying to display a huge range approaching 1,000,000 microns (760K to be exact) on a very small digital or analog display. This squishes all that information in a very small area, changing how it is displayed and hiding some very significant information about what is really happening during the evacuation and decay process.

Decay testing has been used for as long as I can remember to “try” and identify leaks in a vacuum. For me, this process has proven fruitless, frustrating, and more confused than anything because of the science as taught simply never works. In textbooks, a real leak and a virtual leak have very different characteristics making them easy to visually identify while in field practice, it is like trying to identify shades of gray in the dark, you simply cannot tell the difference.

Let There Be Light!

Doing projects like application development with the BluVac+ Professional leads to a lot of testing, and for me, testing means attempting to make the science work. The difference between a real leak and a virtual leak (degassing) should be very apparent and, lo-and-behold, it is if you look at it in the right light. Not only is it apparent, but it is likely the most useful feature of a vacuum gauge aside from total pressure indication (vacuum level) that I have ever found.

Looking at a vacuum in the right light was a simple yet so complex that it was not possible until we attached a smartphone to a vacuum gauge. Unrestricted in user interface design by lights or an analog display, we could for the first time change the way the information was viewed for different elements of the evacuation process. Far from complex, it was as simple as going from logarithmic to a linear display during the decay test.

Decay is shown in a Logarithmic Scale
Same decay is shown on a linear scale graph

Now for you engineers and mathematicians in the group, I don’t expect this to be an epiphany, but for those of us that don’t use logarithmic scales every day or at least me, it certainly was an ah-ha moment. What I had been looking for was hidden, and masked. Not only was this hidden in our design, but literally that of every vacuum gauge that I have ever used.  One small change in how the data was displayed revealed what had been hidden from me for over 30 years, the difference between a real and virtual leak in a system. Because of the logarithmic scale, the decay always visually indicated tendencies of a tightly sealed (but dirty or wet) system that simply needed further evaluation. The tell tail signs of a leak were mathematically hidden from view. Solving one problem, (showing a million of points of information on a single interface) had led to another which was losing characteristics that were critical to solving the actual vs apparent leak problem.

For a service technician, this is a HUGE step forward in evacuation technology because the difference between a virtual leak and a real one are now easily distinguishable in seconds by the means of the app. This means now you know if you simply need more time on the pump, or your system is actually leaking, making a pressure test and a leak test truly necessary. For technicians, this means potentially hundreds of hours saved annually during the installation and service process.

Telling the Difference Between a Leak and “Virtual” Leak

As a system is evacuated, many times the process will stall. The stall happens due to moisture as dehydration is occurring, stalling or slowing the evacuation until the moisture is removed. The question for the technician becomes “is the system simply stalled due to moisture or outgassing, or is there a leak in the system that the pump cannot overcome?”

System stalled in the moisture region. Is it simply wet, outgassing or a mechanical leak?

Now as you will see, on a linear graph this is as black and white as it could ever be. For system (A) below, when the vacuum rig was isolated, you will notice a rapid rise, then a continued rise at a steady rate. This can easily be explained by the fact that with a leak, the air infiltration will happen at a reasonably steady rate provided the system is reasonably below atmospheric pressure. Remember even at 10,000 microns, 99% of the atmosphere has been removed and we are well into a vacuum. Even at these higher levels, a leak can easily be detected.

System (B) shows the characteristics of a virtual leak. There is still a rapid rise at isolation, but now the rise starts to tail off and eventually will rise no more. This characteristic becomes more pronounced at lower vacuum levels when larger hoses are used as considerably more vapors evolve until the system becomes “clean” and degassed. A clean tight system will flatline at isolation because it is clean, dry and tight giving you a good indication of the overall system cleanliness and condition.

This feature of the BluVac+ Professional can easily be accessed at any time by simply isolating the core tools, and tapping the isolate button on the application. The graph will rescale, and in about 30 seconds you will have a definitive answer about what is happening in your system. Leak detection in a vacuum has never been easier. Another great reason to have the BluVac+ Professional in your toolbox.