In the design method the fuel cell power is given, and the rating of the energy storage device(s) is then calculated, based on the energy and power requirements. In this research it is strived to minimize the rating of the energy storage devices, e.g. the battery is designed to have a minimum state-of-charge ofSoCBat,min = 0.2. When the fuel cell power rating is increased the energy requirement of the energy storage device is decreased (until a certain point of the fuel cell power rating, due to the increased energy requirement for the heating). It turns out that the partial cycles are critical for the battery lifetime. The energy storage devices therefore cannot be rated due to the energy and power requirement alone, and the battery lifetime should therefore be included in the procedure of selecting the proper battery rating.
As already mentioned the data sheet of the used battery contains huge uncertain-ties of the cycle-to-failure for partial cycles below a depth-of-discharge ofDoDBat = 0.2. The effect of these shallow cycles should therefore be investigated. The lead-acid battery also suffers from a low lifetime in general. In [54] the cycle-to-failure of a NiMH has been specified toNctf = 8·105 for a depth-of-discharge ofDoDBat = 0.05, and Nctf = 1400 for DoDBat = 1. In [48] the cycles-to-failure have been specified to Nctf = 3000 for a depth-of-discharge of DoDBat = 0.82 and DoDBat = 0.6 for a NiMH and LiIon battery, respectively. For Nctf = 40· 103 the depth-of-discharge is DoDBat = 0.15for the NiMH andDoDBat = 0.1for the LiIon. These values are much better than for the lead-acid, and these batteries should therefore be taken into ac-count. However, these batteries are more expensive than the lead-acid batteries, and therefore cost should also be included.
Another argument of including the cost is because of the power electronics. The power electronics have a high power density and efficiency when compared to most of the other components of the FCSPP. This means that the effect of having a DC/DC converter or not, hardly can be seen. In order to assess the influence of the power electronics, the cost should therefore also be included, e.g. the VA-rating of an inverter 122
8.10. Discussion
is higher when the fuel cell stack or ultracapacitor is connected directly at the bus, than if the bus voltage is kept at a fixed level.
9 Implementation
In this chapter the implementation of the FCSPP in the FC Truck is described. A fuel cell converter that is able to both buck and boost the voltage is used, which means that it also should transit between the two modes in a sufficient manner. Different methods are therefore investigated.
9.1 OVERVIEW
Due to the many partners involved in the consortium, the different tasks were carried out in parallel and not in a sequential order. Therefore the construction of the sys-tem and collection of components were carried out before the analysis in the previous chapter was finish. This means that some of the ratings, e.g. fuel cell power, bat-tery capacity, and bus voltage, are different than the analysis in the previous chapter suggests.
In Figure 9.1 it is seen how the components are placed in the truck. Two inverters are supplying the two motors of PMSM-type. The inverters are connected directly across a 48 V-battery package, but the fuel cell is connected to the battery package through a DC/DC converter. The battery package consists of four series connected lead-acid battery blocks, which are in parallel with four16.2 Vultracapacitor modules.
The fuel cell system is controlled by the vehicle controller which is placed just above the stack. The fuel cell converter and vehicle controller are exchanging data through a RS-232 connection. The fuel cell converter measures the fuel cell voltage and current and transmits them to the vehicle controller. The fuel cell voltage and current are used by the vehicle controller to control the fuel cell stack. The vehicle controller sends enable signals to the fuel cell converter, telling the fuel cell converter if it can draw a current or not. When the fuel cell converter is allowed to draw a current, the vehicle controller also tells the fuel cell converter the maximum current allowed. This is useful during the start-up and shut-down procedure [3]. The inverters are operating alone, i.e. no data are exchanged between the inverters and the vehicle controller or fuel cell converter. It is therefore the job of the fuel cell converter to make sure that the proper amount of fuel cell power is directed to the inverters, and thereby to the motors. The fuel cell converter is also charging the battery package when the fuel cell can provide the necessary power.
Fuel Cell Stack
The blue fuel cell stack is from the company SerenergyR. At the front end it is seen how the heater is connected to the stack. The heater is simply a thin wire connected across the battery terminals. To the heater is also attached a blower. The blower has
9. IMPLEMENTATION
Fuel cell stack
Ultracapacitor modules
Motors Inverters
Battery package Heater
Pipe for exaust gas
Blower Vehicle
controller
Figure 9.1: GMR FC Truck.
two purposes. The first purpose is to blow the hot air into the stack during heating, and the other purpose is to supply the stack cathode side with air, i.e. oxygen, during normal operation of the fuel cell. Right next to the heater and blower a pipe for ex-haust gas is connected. As the temperature for a HTPEM is above100◦Cthe exhaust gas is not water, but steam.
The specifications of the fuel cell stack can be seen in Table 9.1. The specific power of the module is SPF C = PF C,nomM
F C = 131.4 W/kg and the power density is P DF C =
PF c,nom
VF C,nom = 62.2 W/L. These values where used in the previous chapter to calculate the system mass and volume.
Manufacturer SerenergyR
Type Serenus 166 Air C
Nominal power PF C,nom = 920 W Volume VF C = 14.8 L Mass MF C = 7 kg
Table 9.1: Data of the used fuel cell stack.
Inverter
Next to the fuel cell stack the two inverters are placed. The inverters are from SemikronR and they have a built-in Texas InstrumentsR LF2406A DSP, current sen-sors, DC-link voltage sensen-sors, DC-link capacitors, protection against short circuit, over current, over voltage, and over temperature, i.e. they have all the required hardware, and they should therefore only be programmed. The specifications can be seen in Table 9.2.
126