What are Fuel Cell Stacks?

Fuel cell stacks are made up of a number of bipolar plates (also known as fuel cells), which are electrically conducting plates that join the anode of one cell to the cathode of another. Structurally, these are separated by MEAs (membrane electrode assemblies) and bookended with two endplates. The fuel cells each have two cavities for the fuel cell process gases, one for hydrogen and one for air/oxygen, and one internal loop for cooling liquid to pass through in order to maintain the right process temperature for the stack to operate.

Both of these cavities and the internal loop all need to be tested for leakage. A leak of hydrogen to the outside could be catastrophic, as hydrogen is flammable, and a leak from the hydrogen path into the cooling path could corrode the material and/or create bubbles in the cooling channel liquid, causing problems with the pump. A leak in the cooling channel to the outside could cause an electrical shortage, and/or cause the stack to lose cooling capacity.

Depending on the area channel of the leakage, there are different leak rates requirements that apply.

  • Leakage of hydrogen, either to the outside or into the cooling path, should be limited to 10-4 – 10-6 mbar l/s.
  • Leakages from the cooling channel should be limited to 10-4 – 10-5 mbar l/s, as leaks this small will block the cooling liquid from leaking any further.


Why Leak Test with Tracer Gas?

Leak testing using tracer gas is a highly reliable form of leak detection. It gives repeatable and reproducible results, as it is not influenced by outside factors such as temperature and humidity. Crucially, leak testing with a tracer gas allows you to identify the source of a leak, potentially giving you the opportunity to either fix it or reuse the parts of the stack that are not faulty. Leak tests carried out using tracer gas are also traceable to national standards, unlike other varieties of leak test such as bubble testing or residual gas analysis.


Leak Testing Fuel Cell Stacks

Testing for Cooling Liquid Leakage

First, the coolant loop is evacuated of air, filled with helium, sealed, and then placed in a vacuum chamber. Once the vacuum chamber has been evacuated, a mass spectrometer is opened to the chamber. This detects any helium leakage coming from the coolant loop. Finally, after venting the chamber, the helium in the cooling loop can be reclaimed.


Testing for Leakage of Hydrogen to the Outside

The first step for testing hydrogen leakage to the outside is to seal off the hydrogen cavity. The resulting volume must then be evacuated and backfilled with helium. Next, the inlet port must be closed, and the fuel cell plate can then be placed in a vacuum chamber. Similar to testing for cooling liquid leakage, the chamber is evacuated and a mass spectrometer is connected; this enables us to detect any helium leakage from the hydrogen cavity. Once it has been established that a leak exists, a second test (explained below) will show whether the leak is to the cooling channel or to the outside. If not, the chamber can then be vented, and any helium in the hydrogen cavity can be reclaimed.


Testing for Hydrogen Leakage into the Cooling Channel

If a leak was identified in the previous test, the following test will determine whether the leak is into the cooling channel or to the outside. Instead of evacuating the chamber, the cavity is left charged with helium and sealed off, and the cooling channel is then connected to a vacuum pump. The cooling channel is then evacuated to vacuum pressure. The Mass Spectrometer leak detector must then be reattached, and any leakage from the hydrogen cavity into the cooling channel can then be detected. If no leakage is detected during this test, the leak found in the previous test is a hydrogen leak to the outside. The cooling channel can now be vented and any helium within the hydrogen cavity can be reclaimed.


(Ref: “Leak Testing of Components: Fuel Cell Bipolar Plates” by Inficon)