C. Common Testing Procedures

There are many ways to test a fuel cell stack depending upon what sort information is sought. This section will introduce the break-in procedure recommended in order to reach maximum performance and the most common simple tests. Additional tests can be designed by the user in order to provide data to answer specific questions.

After the stack has been assembled for the first time, it is necessary to “break-in” the membranes in order to achieve optimal performance. The point of this process is to hydrate the membranes so as to maximize their ionic conductivity. This is done by introducing humidified reactant flow to the stack. It is recommended that the fuel flow be dead-ended. Remember to occasionally purge at the fuel outlet to drain excess water. Once an Open-Circuit Voltage (OCV) of approximately 0.9 – 1 V/cell is observed, slowly begin increasing the current load on the stack from 0 A up to a maximum of 1 A in steps of 0.1 A. Increment the load once voltage has stabilized from the previous increase. Since the goal is to fully hydrate the membranes, this will be achieved most quickly by pulling the maximum current load that is safe for the stack. A safe level is best determined by the average cell voltage. For brief periods (less than 2 minutes), it is acceptable for the stack to operate with an average cell voltage of 0.4 V/cell. For longer periods of time, a safe minimum average cell voltage is 0.55 V/cell during break-in. Once the stack voltage has stabilized at this level, maintain the current load for a period of two to eight hours. The time required to break-in the stack is dependent upon how high the current load is. One convenient way to determine if break-in is complete is to check the stack voltage ever y few hours at a specific current level, such as 1 A or 1.5 A. If the stack voltage has not changed substantially from the previously measured operating voltage, the break-in process may be considered complete.

It is strongly recommended that the stack be monitored during the break-in period. However, if the stack is to be left unattended, then fuel humidification will need to be discontinued to prevent the anode flow channels from flooding while the stack is unattended. Alternatively, the stack could be run within a continuous flow configuration and humidification may be continued.

After break-in is complete, actual testing may begin. The most common test for a fuel cell or a fuel cell stack is the polarization curve. Polarization curves are also known as V-I or V-j curves as they are plots of the stack voltage against the current load (I) or current density (j). A second common plot that typically accompanies the V-j curve is the power density cur ve, which plots power density (the product of stack voltage and current density) against current density. One method for collecting V-j and power density cur ves is briefly introduced here; for additional information, the reader is encouraged to refer to the multitude of peer-reviewed journals and reference texts such as FuelCell Systems Explained and The FuelCell Handbook.

1. Begin flowing humidified fuel to the stack using either dead-ended or continuous-flow methods as previously described. If desired, use a small fan to actively supply air to the stack. The stack temperature should also be monitored using a thermocouple.
2. Turn on the electronic load and set the current load to 0 A (open circuit). Allow the stack voltage to stabilize and record this voltage. This is the OCV. If the stack is being operated dead-ended, note the hydrogen flow with no load. Under these conditions, if a fuel flow is indicated by the meter/controller, it is most likely due to a hydrogen leak within the system.
3. Increase the current load on the stack to 0.1 A and allow the stack voltage to stabilize. Record the stack voltage and the fuel flow rate. The stack voltage should take about a minute to reach a stable level.

4. Continue to increase the stack current in small increments until the power output of the stack has begun to decrease as current increases or the stack voltage does not stabilize at a particular current load.

5. Using the same current increment, decrease the current load on the stack and record stack voltage as it stabilizes. This will allow for the quantification of any hysteresis effects.

6. Having done a complete current sweep, beginning and ending at 0 A, now the data must be plotted. To calculate the current density, simply take the active area of an individual MEA and divide the current load by this value. For example, a 1 A load on a 10 cm² MEA is a current density of 100 mA/cm². Plot these current densities on the x-axis. Plot the stack voltage on the y-axis. Take the product of the current density and stack voltage to calculate the stack power density, which will have units of W/cm² or mW/cm². It is common to plot the power density curve on the same graph as the V-j curve, using a second y-axis so that the stack current and power density are properly scaled.
7. Several parameters can be varied to yield a significant number of V-j curves for basic analysis—including running the stack with continuous fuel flow versus dead-ended fuel flow, running the stack with humidified versus dry fuel, testing the stack with a passive air supply versus a forced air supply, varying the fuel flow rate (continuous flow) or fuel pressure (dead-ended), and varying stack temperature.