When choosing an electronic load for testing fuel cells you need to first determine what kind of electronic load you are going to use. There are several categories of Electronic Loads: The older, more standard, Electronic Loads (Transistor Types). The newer Electronic Loads (FET High power types). Or the latest, most reliable, type of Electronic Load (Low Voltage Low ON-Resistance FET types) that work very well in testing fuel cells and batteries. Of course you could always do it the old way with a bunch of power resistors, but then the results won't be as good. There is also what is called a switching electronic load. This type of electronic load can cause some real problems for a device under test. I suggest strongly that you never use a switching electronic load. Picking the correct electronic load sets the stage for all the other processing required in testing a fuel cell (fuel cell refers to all types of devices). So why use an electronic load? Well that's easy to answer. An electronic load acts just like a constant current power resistor with an adjustable range of current. That takes at least one of the variables out of the equation, the current of the fuel cell. The electronic load(s) holds the current constant (try doing this with a power resistor). Now you are left with just the voltage varying from the fuel cell. Therefore, for production testing applications you are now left with only a passed/failed type of operation. The process in choosing an electronic load is simple. First you need to know the cell voltage or cell stack voltage of the fuel cell. Then you need to know the current of the cell and/or cell stack. All electronic loads are rated in watts (Volts x Current = WATTS), most at room temperature. If you plan to do testing at other than room temperature, read the electronic load's specifications. Above standard room temperature most loads will lower the power they can dissipate. The next thing you will be looking for is the electronic load's top voltage it will work up to and the bottom voltage it will work down to. This process can be tricky. As the voltage operation goes down so does the load's correct current operation. Lots of electronic loads only work down to 0.7 or 3 volts. Below that the electronic load might not work properly, or at all. If you are testing a single fuel cell you will need a load that works below the 0.7 volt range. A suggestion to get around this problem is to put a power supply in series with the load thereby boosting up the voltage of the device you are trying to test. Of course there are some drawbacks, voltage instability of the power supply, instability of the electronic load, and noise from the power supply, over heating the UUT, changing the chemistry of the UUT, high frequency current that you can't see in the testing, dynamic instability, reduced slew rate, current feedback, this is just to mention a few problems, there are a lot more and unless you are a power supply test engineer you would never know them all. The instability of the power supply line and load regulation can be several hundred millivolts, more than the voltage of a single cell of a fuel cell. If you are using a power supply to boost up the voltage you will need to take lots of measurements at different points in the circuit to get the correct readings of the fuel cell. When doing this type of testing never use a switching power supply as switching power supplies have lots of noise and current spike problems that you may not be able to see with standard test equipment and this may cause an erratic reading of the fuel cell. There is also the added problem with a power supply in series dynamic test. It is almost imposable to get good readings because of the added inductance and capacitance of the power supply, you have a lot of added unknown elements added to the circuit. When testing a fuel cell with a power supply in conjunction with an electronic load the line and load regulation of the power supply will always be the two most important factors. The line regulation is the AC input voltage to the power supply. That input has a direct effect on the output voltage and current output of the power supply. The load regulation is the amount of current you are drawing through the power supply output. That current can show up as a voltage increase or decrease in the output voltage as the current changes through the fuel cell. These changes can show up as a few millivolts to hundreds of millivolts over the operating current range of the power supply. Always measure the specifications of each power supply yourself. Don't ever take the word of the manufacturer as they sometimes make the supply's specifications look better than they really are. Power supplies have another hidden problem, sense leads. Sense leads make up the voltage drop on the wires that are hooked to the power supply. Should you use the sense leads or not? If you do use sense leads where do you hook them? Normally you should hook the sense leads to the end of the power wires of the device they are powering. But, for a fuel cell you should hook them to the output terminals of the power supply. This will help stabilize the power supply and give optimal performance from the power supply when testing a fuel cell. Do not hook them anywhere else as this will lead to false readings in the measurement process of the fuel cell. Keep the sense leads as short as possible and keep them away from any power line noise or high voltage noise generating devices. Using a power supply in series with an electronic load is never a good idea, use this method only as a last resort if there is no other way of doing your testing. I have only covered some of the more common problems that can happen using this method. You should always consult a power supply test engineer before doing this type of testing. Now that you know a little about using power supplies with electronic loads let's continue with the information you need to know about the electronic load itself. The biggest problem with an electronic load is leakage current, or idle current. There is always leakage current. This can vary from a few mills to hundreds of mills of current and most manufacturers will not specify the amount. This may, or may not, be important to your testing but even if it is not you should ask the manufacturer this specification. The leakage current can change as the voltage applied changes causing the leakage current not to be linear. And in some cases the leakage current can change with the amount of current that is drawn by the electronic load. To get good measurement results for your fuel cell you always want a load that has low leakage current changes. Temperature stability is going to be where most testing is going to fail and a lot of this can be avoided by some good engineering practices. Don't use a 3000-watt electronic load to test a 3-watt circuit. A lot of the stability of the electronic loads are based on just how many power devices are used in the loads power circuits. The more power devices you have, the more leakage current and temperature instability you are going to have. You should use loads that have about 20 to 50% more power than you will need. This is the optimal range for best stability of an electronic load. Old load designs are another type of problem. Old types, or more standard electronic loads, dissipate power using transistors. This type gives you very good stability for long wire connections to the source to be tested. For the most part the slew rate of this type of load is very slow and the lowest voltage that can be tested is about 2.5 volts. Remember, below the bottom rated voltage you can have erratic problems with the load or the current readings may be incorrect. You should always check this specification. Trying to use a load below it's voltage specification can be risky. You should always check the current reading with another type of current measuring device to make sure the electronic load is working correctly. The other type of electronic load is a FET type. This type can test lower voltage type sources and has faster slew rates. It does have some drawbacks in that long leads to the load can cause instability, or oscillation, of the FETs in the power stage. You will also see added capacitance at the load side when testing the power source. To add operational stability to the load most manufacturers of this type of load add capacitors inside of there loads. In general their current leakage is lower and their temperature stability is better. You will be faced with another problem if you try to remotely control the electronic load. You may encounter ground loops from the load to the control card, again causing all types of instability. The best electronic loads offer isolated control sections, or options, for isolation control of the load. Typical control ranges for the analog input is 0-5 volts and 0-10 volts. Check to see if the slew rate of the voltage control also controls the slew of the load. In some cases it does not control the slew of the load and is used only as a voltage to current reference. The load will be working in constant current mode and if you require constant power or constant voltage you will need to write some software to do these kinds of control function. This may require a multifunction data acquisition control card. This is not a simple kind of program. In most cases it takes several weeks to produce good code for the control software to work with the electronic load. If you are going to use an IEEE-488 interface be prepared for very slow responding software. The IEEE-488 is almost unusable for any type of high speed production testing. In testing a fuel cell you may want to know more than just the Volts and Amps of an electronic load. Volts and Amps don't tell you enough about the load or it's operation. You also need to know the ON-RESISTANCE of the electronic load. This parameter tells you the lowest voltage and the highest current that you can reach with the electronic load. It tells you something about the loads circuit impedance and how the load and fuel cell may react to each other. It also tells you a little about how the load may drift with temperature. The higher the ON-Resistance the more the load will drift with temperature. The ON-Resistance of the load should be about 10 times less than the impedance of the fuel cell being tested. The spare resistance will be eaten up by other circuits you need to add at a later time. Wiring is another issue most engineers don't take into consideration when testing high power fuel cells. If you are only testing 1 to 5 amps, that's not much current. But when you get into that 100 to 250 amp range that is a lot of current. A few milliohms of resistance in the wire and connections makes a lot of difference in the voltage drop to the electronic load. It also is a factor in how the electronic load will view the current that it is controlling. The voltage at the load may be very different than the voltage at the fuel cell. Just how large a wire should be used depends on the current you want to draw and how much voltage you are willing to drop across the wire. In the case of 250Amps you should be using 00# to 0000# wire, which may not even be large enough, depending on how long the wire run is. You can also use a current shunt to measure the fuel cell current. For example, to measure 100 amps of current the current shunt has to drop voltage since the current is reading out in volts and/or mV per amps. Lets say that the shunt is 0.001 V/A, that's 1mV per amp. 100 amps x 0.001 = 0.1 volts. If your fuel cell is a 0.3 volt unit you just lost 1/3 of the voltage you are using for testing. That's not much voltage to go to the electronic load. Your wiring could drop the rest of the voltage and you could get 0 volts by the time it gets to the electronic load. You always want as much voltage at the electronic load as possible when testing a fuel cell. In general, the more voltage the more current range you will have at these low voltages. This just touches on some of the more common problems that engineers encounter and should be aware of in selecting an electronic load. Fuel Cells and Batteries require special electronic loads designed for only that application. Most standard electronic loads don't work well in these applications. Don't expect to go to a website and find a ready made electronic load that works only with fuel cells. You will be required to build your own test stations for each different type of fuel cell application as each application has it's own special problems that have to be taken into account. Executive Engineering makes electronic load blocks just for this purpose. Each block is a miniature module that can be connected in different ways to produce an electronic load that fits your projects needs. These blocks can be connected to each other to increase the current or connected in series for special charging and discharging circuits. Executive Engineering has made it very easy to build a customized electronic load by just wiring modules together.
|
|
| |