Vacaero

Manufacturers of heat treating and brazing vacuum furnaces and controls, complete hot zone and vacuum furnace retrofits, thermal spray coatings, plasma, HVOF and paint coating services.

VAC AERO Service Experience Trust
Canada FrançaisAccessibility |

Finding and Fixing Vacuum Leaks

by

Finding leaks in vacuum furnaces is a task that few of us cherish but all of us know is important and necessary. Leaks are a problem experienced by almost every vacuum user. Leaks can occur suddenly or develop slowly over time but in either case, they are damaging both to product quality and to furnace internal components. In extreme cases, the problem is obvious: the furnace will not pump down and/or the hot zone (or heating elements) shows obvious signs of attack. Tiny leaks, however, are more common often going undetected because of the pumping systems ability to overcome them. However, even small leaks can cause continuous and sometimes catastrophic damage. Thus, routine leak checking and leak repair should be a part of any good vacuum furnace maintenance program.

Like any job, leak detection can be made easier by having the right tools, looking in the right places, having patience and using a great deal of common sense. Leaks can be inherent in the materials of construction, created during the manufacturing process, be introduced during maintenance or repair, or occur over time due to corrosion, wear, fatigue or stress. The question is: “Can the system tolerate the leak?” In other words, can the process and equipment survive the consequences of the leak? The answer is almost always, “No.”

The most obvious place to look for a leak is in and around the last place that was repaired or serviced. Begin by asking the question, “What was the last area worked on or modified? ” (e.g. thermocouples, power feedthroughs, door “O” rings, etc.). It is incredible how many leaks are found and corrected by simply asking questions before one begins leak detection in earnest.

Many different techniques can be used to find leaks. Below are two of the most common.

The Solvent Method

The use of a solvent test method is simple but effective for locating gross leaks, that is, leaks considered in the intermediate to large size range. If a thermocouple gauge is connected on the pump side of the system such that it reads manifold pressure and the system can be evacuated to a range of at least 200 microns (0.2 torr) then a solvent such as acetone (preferred) or alcohol can be carefully sprayed on a suspect area. Any change in vacuum level inside the chamber, observed on the vacuum gauge, signifies a leak in that area.

With this method, one must be careful to allow enough time (up to 20 seconds) for a pressure increase to take place. This procedure is more sensitive at lower pressures. Solvent checking is typically used to fix gross leaks so that the system to be evacuated into the range where a mass spectrometer instrument can be used to check for smaller leaks. Be sure to observe all required safety precautions when using hazardous solvents, including proper ventilation and spill containment. Remember also that in many cases these solvents will remove paint!

If a leak is located, a temporary sealant such as Glyptal® red alkyd lacquer (Glyptal Inc., Chelsea, Mass., www.glyptal.com), Kinseal clear vacuum sealant (Kinney Vacuum Div., Tuthill Corp., Springfield, Mo., www.kinneyvacuum.com), vacuum seal putty or wax can be used to patch the area and allow leak checking to continue. A common mistake is to forget to permanently fix the problem after testing is completed. Remember to permanently fix any temporary patch in a vacuum furnace which will be running at a positive pressure as the gas pressure will open up the leak again. In these instances, the component must be taken apart and fixed.

The Helium Mass Spectrometer Method

A helium mass spectrometer (Fig. 1) is a highly accurate instrument for locating leaks, especially those in hard-to-reach areas. In some instances, it is necessary to “bag” or isolate a specific area on the furnace and inject helium into the contained space. This method is sensitive enough even to locate leaks in moving or transition (vacuum to pressure) seals.

Figure 1 [1] Portable Helium Mass Spectrometer with Remote Control Capability (Photograph Courtesy of Agilent Technologies)
Figure 1 [1] Portable Helium Mass Spectrometer with Remote Control Capability (Photograph Courtesy of Agilent Technologies)

A mass spectrometer can detect extremely small amounts of helium. Helium is the tracer gas of choice because it is inert, nontoxic, relatively inexpensive (in small quantities), and not easily absorbed. Helium also easily flows through small leaks and has only a trace presence in the air (usually 5 ppm).

The theory of operation is as follows: when the gas enters the spectrometer tube it is ionized and accelerated. The resultant high-speed charged particles are then exposed to a magnetic field perpendicular to their direction of motion. The resultant force is perpendicular to both the velocity vector and magnetic field and causes the particles to follow a curved path, the radius of which depends on the mass of the particle, thus, allowing separation of the particle stream into different ions. A properly positioned collector plate (ion detector) enables the concentration of any gas to be very accurately measured. Every electron given up by the collector plate equates to the presence of one helium ion. The amount of helium collected is then converted to a leak rate.

The following technique [2] is a practical guide to using a helium leak detector. Be sure that the system you are using is properly calibrated and test that the system is working. A typical procedure for setting up the leak detector (for a Varian Model 858 Turbo Leak Detector) is as follows. This procedure assumes a calibrated leak device is installed. Specifically:

  1. Block the leak detector test port with the supplied plug.
  2. Start the leak detector by turning on the main power switch.
  3. The unit is ready when the “Hi VAC OK”, “Fil” and “Ready Turbo” lights are illuminated.
  4. Press the “start” button and wait for the “Test” button to illuminate (both are typically located on the front panel).
  5. Set the range to 9 (10-9 torr) and wait for the Leak Rate bar graph to stabilize.
  6. Adjust the Zero knobs (coarse and fine), if needed so that the Leak Rate bar graph displays a reading and is not illuminating the “UNDER” light.
  7. Press the “Vent” button, located on the front panel of the leak detector, and remove the plug.
  8. Open the valve on the calibrated leak and press the “Start” button.
  9. With the valve on the calibrated leak closed, ensure that the Range is still set to 9 (10-9 torr).
  10. The Leak Rate bar graph should be displaying a value similar to what was observed in step 6.
  11. This is the Zero calibration.
  12. Open the valve on the calibrated leak and set the Range to 6 (if the “Hold” button illuminates after opening the value on the calibrated leak, repress the “Start” button).
  13. The Leak Rate bar graph should be reading between 2 and 4.
  14. This is the Span calibration.
  15. If the leak detector passes the Zero and Span calibration it is ready for use.
  16. Remove the calibrated leak by pressing the “Vent” button, closing the valve on the calibrated leak and remove it from the detector.
  17. Install the vacuum test line to the furnace test port (typically located in the vacuum pumping port).
  18. You are now ready to commence leak detection (see below). The Range should be set to 9 (10-9 torr) and the “Start” button pressed.
  19. If the detector fails the Zero and Span calibration, perform the tuning procedure described in the vendor’s operations manual.

Begin by pumping down the chamber (including the use of the diffusion pump if available) and be sure you have reached the lowest vacuum level possible. Next, check valve in the leak detector and note the background leak indication. It must be less than the standard leak, preferably a decade lower than the standard. High background indications suggest gross leaks in the 10-4 or 10-3 cc/second range, so these leaks need to be repaired in order to get the background down to a level where helium mass spectrometry is effective.  Adjust the helium flow on the wand to a low flow setting (allowing the helium to contact a wet part of the skin or use a liquid and adjust the flow to several bubbles per second.

Start at the top of the vacuum system and work down, checking the following areas:

  1. Power feedthroughs
  2. Thermocouples ports- check the sheath gland and pipe threads – don’t forget to bleed helium at the connector to verify that there is no leak down the inside of the sheath.
  3. Door seals
  4. Viewport seals
  5. All tapered pipe thread connections
  6. Vacuum/pressure switches (note: remove the switch covers and bleed helium in)
  7. Electrical feedthrough on quench blower motor (if applicable)
  8. Atmospheric glands on internal cooling loops
  9. Nitrogen/inert gas bleed valves (this will require opening the upstream connection and introducing helium into the valve)
  10. Any Dresser-type couplings
  11. Oil drain/fill connection on the diffusion pump (note: flood bottom of the pump to see if there are leaks on the boiler plate)
  12. Cold cap glands on the diffusion pump (if applicable)
  13. Butterfly valve shafts
  14. High vacuum valve piston rod seal (may require introducing helium into the lower port of the high vacuum cylinder)
  15. Shaft seals on oil agitators.  If agitators have “canned” motors, check electrical feedthrough on motor
  16. Oil to water heat exchangers used on oil quench furnaces. This would require blowing out the water on the tube side and introducing helium- a very involved procedure
  17. Mechanical seal on quench oil circulation pump
  18. Mechanical seal, end plate, and oil line leaks on the roughing pump
  19. Mechanical seal, end plate, oil fill/drain leaks on the blower

Additional leak checking hints include:

  • On old furnaces, especially those that use untreated water for cooling, check tube penetrations from the outer to the inner shell as these often corrode through and leak water. Gross leaks will be obvious but pinhole leaks will allow ice to form on the vacuum side. Since ice has a much lower vapor pressure than liquid water, it delays pumping below 0.067 mbar (50 microns). Very small leaks will not show up as water in the mechanical pump. Gross water leaks result in a “milky” appearing pump fluid.
  • Helium “drift” is the major problem in locating the site of the actual leak. The problem is the use of too large a flow of helium, which results in a leak indication that is not in the location probed. When you are on top of the leak, the response time will be very quick, in the order of a second or two, and you should be able to repeat the leak indication at least three times successively at the same location.
  • If no leaks are found, note the vacuum level and slowly open the needle valve on the main chamber to give a small upscale deflection on the vacuum level. Now introduce a small amount of helium into the metering valve. You should see a large leak indication within 1 – 2 seconds. Then close the metering valve.
  • Probe with helium into the gas ballast check valve on the mechanical pump. Close the port to seal it off and then quickly open and close the gas ballast ball valve to introduce a small amount of helium into the pump. It will “back diffuse” into the blower and allow you to see a large leak indication quickly. Leaks that remain could be due to water leakage on the furnace shell or gas cooling finned heat exchangers. The gas cooling heat exchangers will have to be blown out to eliminate water, and then helium introduced using a large helium tank with regulator. Water leaks on the diffusion pump cold cap are also common. Check this by blowing out the cap water line and introducing helium.
  • With all leaks repaired, a background leak indication in the low 10-9 cc/second range should be achievable. You may have to vent and re-pump several times to get a low enough background.
  • At the conclusion of leak check, get in the habit of valving off the leak detector so that it won’t be dumped to air on the next roughing cycle.

Mass Spectrometer Leak Testing Tips

Mass spectrometer leak testing requires that the unit be exposed to helium leaking for only 3 to 4 seconds. However, as with the solvent test method, a dwell or lag time between test areas is needed to prevent false readings. So being too quick (overly aggressive) in finding leaks using helium is a bad practice often resulting in wasting a great deal of time repairing areas which were not leaking in the first place. Avoid the tendency to “move on” before allowing adequate time for the helium to migrate from outside to the detector.

In utilizing helium, checking should always begin at the bottom of a pressurized system and at the top of an evacuated system. When dealing with moving or transition seals (e.g. vacuum seals that must also withstand pressurization) it may be necessary to “bag” the area under investigation, using a plastic bag and tape to seal off a component and then inject helium into the closed area.

Summing Up…

People performing leak detection should be properly trained, given the right tools and the budget necessary to do their jobs. Leak repair is part of a comprehensive vacuum furnace planned preventative maintenance program. Proper care of pumps, replacement of “O” rings as they age, cleaning of flange sealing surfaces (particularly loading door seals) and regular inspection of vacuum thermocouple fittings and power feedthroughs will help prevent the majority of leaks. In addition, daily leak checks or continuous monitoring of vacuum levels during processing can also help to identify potential problems before they develop into major repairs.

References

  1. Herring, Daniel H., Vacuum Heat Treatment, BNP Media Group II, 2012.
  2. Mr. Geoff Humberstone, Metallurgical High Vacuum Corporation (www.methivac.com), editorial review.
  3. Practical Vacuum Systems Design, The Boeing Company.
  4. Leak Detection, Applications & Techniques, Agilent Technologies.
  5. Grann, James, Understanding the Difference Between Linear and Non-Linear Leak Rates, Symposium on Vacuum Furnace Maintenance, ASM International, October 2007.
  6. Herring, D. H., The Why, When and How of Leak Checking a Vacuum Furnace, Heat Treating Progress, September/October 2003.
  7. The Nature of Vacuum, SECO/WARWICK Corporation.
  8. Brunner, William F., Vacuum Leak Detection, American Vacuum Society, 1981