The first step in the electrical system design was to ...



Technical Data Package

Feb. 24th, 2006

Group Members: Sean Ashman, Chad Byler, Dennis Farley, Brian Holzberger, Shan Hu, Corey Reynolds, Dan Upton, Steve Yang

Introduction

The goal of this project is to create an apparatus capable of testing the performance of hydrogen fuel cells for the NanoPower Research Laboratory (NPRL) located within the science department of Rochester Institute of Technology. The hydrogen fuel cell has recently become a popular topic of discussion as oil production is disrupted by military conflict abroad and natural disasters domestically. How a fuel cell works, however, is not general knowledge. The first task of this project was then to discover how a fuel cell works. Second, then was to realize our customer, the NanoPower Research Lab (NPRL), is testing an experimental fuel cell setup, that is not precisely like a typical cell. The NPRL manufactures their own nanotube catalyst layer internally, a critical component of the hydrogen proton exchange membrane (PEM) fuel cell. It is the performance of this nanotube catalyst layer compared to more traditional catalyst layers used in fuel cells that they wish to test. The customer is also interesting in that they are an expert in carbon nanotubes but are new to fuel cell research. Consequentially they are not sure of the best practices or the standard test procedures used in other fuel cell labs. The customer’s original need statement was, “If you can duplicate the capabilities of the test cell in the Mechanical Engineering Lab, you’ve succeeded.” However, after an answer of $1,500 for our budget, but an answer of $29,950 for the TVN Systems RU-2100 that sits in the Mechanical Engineering test lab, we realized what the project was. The needs of our customer NPRL can be divided into three general categories, the Mechanical Assembly section, the Interface and Control section and the Hydrogen and Oxygen Gas Supply section. The primary need relating to the Mechanical Assembly subsection is the assembly of the fuel cell itself. The mechanical assembly must supply enough compressive force to contain the various layers that make up the fuel cell and the hydrogen and oxygen gas being supplied to the fuel cell. The assembly must also be able to heat the fuel cell to a temperature up to 80°C, and maintain desired levels of testing temperatures. The principal need under the Interface and Control section deals with the environmental parameters within the system. The environmental parameters pressure, temperature and humidity must all be monitored and recorded. This data also needs to be put into a format compatible with Labview, as this was the customer’s software package of choice. The chief need of the Hydrogen and Oxygen Supply section is to be able to supply Hydrogen and Oxygen at selectable pressures and humidities to the system.

All those viewing this project, must have at least a passing familiarity with the operation of a proton exchange membrane (PEM) fuel cell, so here it is.

[pic]

Basic Operation Diagram of a Proton Exchange Membrane

The proton exchange membrane fuel cell uses one of the simplest reactions of any fuel cell. Within a standard fuel cell, there are four basic components; the anode, the cathode, the proton exchange membrane and the catalyst. The anode is the negative post of the fuel cell. It performs several functions. It conducts the electrons that are freed from the hydrogen molecules so that they can be used in an external circuit and it has channels etched into it that disperse the hydrogen gas equally over the surface of the catalyst. The cathode is the positive post of the fuel cell. Like the anode, the cathode also has channels etched into it, but these channels distribute the oxygen to the surface of the catalyst instead of the hydrogen. The cathode also conducts the electrons back from the external circuit to the catalyst, where they can recombine with the hydrogen ions and oxygen to form water. The proton exchange membrane (PEM) is a specially treated material that only conducts positively charged ions. The PEM blocks electrons. The catalyst is a special material that facilitates the reaction of oxygen and hydrogen. It is usually made of platinum powder very thinly coated onto carbon paper or cloth. The catalyst is rough and porous so that the maximum surface area of the platinum can be exposed to the hydrogen or oxygen. The platinum-coated side of the catalyst faces the PEM. The NPRL utilized nanotubes in the construction of their catalyst layers. On the anode side, Hydrogen gas is forced through the catalyst by pressure. When an H2 molecule comes in contact with the platinum on the catalyst, it splits into two H+ ions and two electrons. The electrons are conducted through the anode, where they make their way through the external circuit (doing useful work) and return to the cathode side of the fuel cell. While all this is happening on the anode side, the cathode side of the fuel cell is also busy. Oxygen gas is being forced through the catalyst, where it forms two oxygen atoms. Each of these atoms has a strong negative charge. This negative charge attracts the two H+ ions through the membrane, where they combine with an oxygen atom and two of the electrons from the external circuit to form a water molecule H2O.

This technical report divides the material covered during the winter quarter of 2006 into the following topics.

• Layering of Fuel Cell

• Mechanical Assembly Process

• Heating of the Fuel Cell

• Humidification of gas

• Exhaust/Back Pressure Control

• Electrical Sensors and Power Supply

• Heating elements

• Layout of program logic

• Data Acquisition

• Budget

Gas Diffusion Plate/Electrode Assembly

The testing of the PEM fuel cell requires that a good electrical connection be made to the carbon nano-tube catalyst. The difficulty in this is that in order for the cell to work, H2 and O2 need to be able to come in contact with the nano-tube layers. This requires that the electrode also have means for the gases to pass through them as well as provide electrical contact to the nano-tube layers. The customer also expressed interest in being able to change the reaction area of the nano-tube layers. This required either adjustable electrodes or multiple electrodes. Initially it was discussed to use an electrode that would essentially be a washer. This would allow a fast, easy, cheap, and accurate method for producing an electrode that would be able to handle the current generated by the fuel cell, control the reaction area of the fuel cell, and maximize the gas contact with the nano-tube layers. After discussing this design idea with the customer, we were informed that due to the nature of the nano-tube technology being used, the electrical resistance in the nano-tube layer was higher than that of conventional electrode layers. This required us to find a method that would maximize both the electrode contact and the gas contact with the nano-tube over the same area.

The solution came in the form of a mesh. It would provide a good electrical contact over a large area as well as allow gas to pass easily through and reducing low concentration of the gas on the layers (area where the electrode would block gas from contacting the catalyst layer). Using a mesh type electrode provided the surface area needed by both the electrical aspect as well as the fluid aspect of the project, but the mechanical pressure exerted by the electrode to provide good electrical contact needed to be addressed.

Multiple ideas for creating a good positive pressure on the fuel cell stack were discussed. Some included springs that would provide constant pressure on the nano-tube layers, others simply used mechanical pressure from a power screw or other device to maintain constant pressure. Knowing that the plates were required to be under pressure to create a tight seal to the nano-tube layers but not make contact to the PEM fostered a solution that utilized a raised portion on the electrode that would be the contact surface for electric current to flow as well as have a mesh designed into it so as to allow gas to flow through as is seen in the picture below.

[pic]

Now that a design for the electrode was decided, a method of mounting the electrode to the clamp assembly needed to be assessed. Since the clamp assembly is primarily metal, an electrically insulating material would have to be inserted between the mechanical press and the electrode. This insulator would also have to allow gas to flow through so it could go through the electrode and it had to withstand the pressure of the mechanical press. Because of these conditions, a composite material would be used as it provided good electrical insulation and could withstand the compression loads of the mechanical press. The insulator would also allow for quick tube fittings to be installed that would carry the gases to and from the fuel cell.

[pic]

The meshed electrode is mounted to the insulator with ¼-20 flat head cap screws. A short rod that is threaded on the inside and outside is inserted through the back side of the insulator and in conjuncture with a wire lead allows the current to be carried to the terminals of the electrical sensors. A thin self adhesive seal is attached to one electrode of a stack to help seal the gases from escaping.

As the fuel cell stack is intended to be assembled off from the mechanical press but be easily mounted to the mechanical assembly, a method of assembling the two sides of the fuel cell stack together was needed. Short rods with spring loaded balls were used. They allow both alignment of the sides of the stack and the temporary assembly before the stack is inserted into the press.

[pic]

Mechanical Assembly

One of the problems that is encountered by the customer is that the fuel cell is assembled by hand and held together with 4 bolts. When the cell is bolted together there is no way to easily achieve a repeatable pressure. The process is time consuming and can possibly lead to hazardous situations if the assembly is not bolted together properly. If it is assembled improperly Hydrogen or Oxygen can escape out of the assembly and present a fire hazard.

The system designed must be able to provide a repeatable mechanical pressure applied to the assembly. The system must also prevent possible misalignments that allow gas to escape and to seal the system to prevent this outward flow. The assembly should also be able to quickly and easily assemble the fuel cell.

Based on these requirements we were able to come up with an initial mechanical assembly design. This mechanical assembly included a rotating backing plate that allowed the fuel cell to be assembled on a horizontal surface that then locked the fuel cell into position, using pegs that replaced the 4 bolts, and allowed it to be rotated to a vertical position. Once in a vertical position the assembly can be compressed using another plate attached to a power screw for compression. The compression plate is guided by a machined slot that prevents rotation.

[pic]

The power screw used needs to be able to handle a max internal fuel cell pressure of 60 PSI, including a factor of safety of 2 that is 120 PSI that needs to be handled. The initial fuel cell design had internal fuel cell dimensions of one and 5/8 inch square. Using these dimensions and power screw specifications for a ½”-10 power screw we can calculate the torque required to compress the fuel cell.

|Internal Pressure |120 |psi |

|Surface Area |2.64 |in^2 |

|Max Force |316.8 |lb |

|tan(λ) must be less than the coefficient of friction in order to be positive locking |

|Screw Type: 1/2" - 10 |  |

|Lead |L= |0.1 |in |

|Root Dia. |Dr= |0.45 |in |

|tan(λ)=L/(π*Dr) |  |  |  |

|tan(λ)= |0.071 |  |  |

|m = |0.3 |Coefficient of friction steel to steel |

|tan(λ) < m, therefore the screw is self locking |

|T=F*Dr/2*((L+π*m*Dr)/(π*Dr-m*L) |

|T = |26.999 |Lb/in |  |

|T = |2.25 |Lb/ft |  |

This means that it will take 2 and ¼ lb/ft of torque to compress the system in a worst case scenario. This is a relatively small torque and means that this is a good power screw for our system if there is enough space to keep it in. The design also requires the power screw to be self locking and using the screw geometry we are well under the friction limit.

There are other advantages that this fuel cell design provides that will allow us to better control the fuel cell environmental conditions. The sponsor wanted to have a way to control the temperature in a range of 20oC to 80oC. In order to do this the backing plates can have water jackets machined into them. This is achieved by moving the 4 pegs to the corners of the plates. This provides adequate clearance to drill 3 holes in the plate, such that it will allow the flow of water. One of the holes will be plugged to prevent flow out.

[pic]

Once the ability to heat the system was added there was a strong desire to enclose the system to contain as much heat as possible. In order to do this removable enclosure plates were added. These removable enclosure plates will also help to contain gases should something horrible happen like a fire breaking out.

[pic][pic]

(Pictures shown without removable enclosure plates.)

[pic][pic]

(Pictures shown with removable enclosure plates.)

The original design included slots machined into the side plates. This was changed in favor of a plate that moves along 4 ground aluminum rods using thrust bearings. The design was changed in order to achieve a better sealing surface between the enclosure and the compression plate.

Plain Bearings are needed in two areas. We need them for the rotating assembly and for the power screw connection. All bearings used in this design are oil impregnated bronze as they are a highly cost effective option.

The power screw has a snap ring that is attached on the backside of the compression plate in a recessed slot. The snap ring groove will be machined into a purchased power screw. The connections for the pressurized gas lines come through the same slot making for a tight packaging situation.

[pic]

The customer made a request to have maximum internal dimensions of a 2 inch by 2 inch square. This has little impact on the overall design except that the power screw calculations need to be redone.

|Internal Pressure |120 |psi |

|Surface Area |4 |in^2 |

|Max Force |480 |lb |

|tan(λ) must be less than the coefficient of friction in order to be positive locking |

|Screw Type: 1/2" - 10 |  |

|Lead |L= |0.1 |in |

|Root Dia. |Dr= |0.45 |in |

|m = |0.3 |Coefficient of friction steel to steel |

|T=F*Dr/2*((L+π*m*Dr)/(π*Dr-m*L) |

|  |  |  |  |

|T = |40.908 |Lb/in |  |

|T = |3.409 |Lb/ft |  |

Because the screw geometry does not change the self locking calculation remained the same. The torque input required is still a relatively low torque and does not require a redesign for a larger power screw.

Fuel Cell Heating System

The objective of the heating system is to provide a controllable and sustainable temperature inside the fuel cell assembly. This is important because several types of losses in the cell are highly dependent upon temperature. By varying the temperature of the cell and holding it at a fixed temperature results can be compared and losses interpolated.

Several design concepts were compared based on relative advantages and disadvantages. The two concepts that were considered most reasonable and practical were a fluid flow heating system and a resistive heating system. The main issue that ended up being the deciding factor between the two had to do with the fundamental operation of the resistive heating method. The resistive heaters convert electrical energy into heat energy by way of running current through a resistant medium. However, this not only creates heat energy but also a rather large magnetic field. The fundamental operation of the fuel cell requires a flow of electrons and we were not sure what kind of interference a moderate to large magnetic field in such close proximity to the cell would have on performance.

Settling on the heated water flow system we pressed forward with the design process. The most suitable method for heating the water was still determined to be a resistive heater but by using water an in intermediary agent we were able to move the heaters a comfortable distance away from the cell. The compression plates that press on the cell from both sides are a natural and easy way to implement our solution for heat delivery to the assembly. The resistive heaters heat water in an external reservoir. The water is then pumped by an inline water pump to the compression plates where it transfers some of its heat to the cell creating an in-cell temperature distribution seen in the ANSYS figure below (assuming no insulation). The water flows through the compression plates and circulates back to the reservoir where it picks up more heat. Insulation will be employed in the final design in order to keep as much heat energy as possible inside the system.

[pic]

Figure 1. – Cathode side of cell. Assuming uniform temperature across

compression plate and no insulation around cell.

Given a resistive heater of 100W the following calculation serves as a guidline for time to heat water in a worst case scenario 20°C (room temperature) to 80°C (highest allowable test temperature).

[pic]

Flow System

The initial design objective for the flow system was supposed to incorporate an electrolysis system that would make on demand hydrogen and oxygen gas. This was an adaptation of the system currently in use in the NPRL. Several design ideas were conceptualized for pressurization of an electrolysis system.

1. Basic Piston in Pressure Vessel

2. External pressure vessel with compressor

3. In-system pressure vessel with compressor

The piston in a pressure vessel concept presented too many issues with sealing and

temperature increase to be considered seriously. The in-system pressure vessel would have consisted of the two electrolysis tubes capped and hooked with a T-connector to an external compressor. The in-system pressure vessel design and its successor, the external pressure vessel, were the main design concepts for the flow system for the majority of the design process. The external pressure vessel simply relied on a third external tank to the electrolysis tubes to act as an intermediary for pressurization. This design reduced gas loss issues in the form of diffusion from the increase in pressure as well as providing a more equalized pressure application between the electrolysis tubes.

When the design of the external pressure vessel was presented to the customer it was decided that the system was too expensive and time consuming, not allowing enough expansion for future development. It was dropped in favor of a compressed gas cylinder system with simpler heated tubes which humidify the incoming gas to a set level based on the temperature of the water which is explained next.

Humidifiers

As the fuel cell’s performance can vary with the humidity of the two gasses. The customer asked that the test stand be capable of running tests at between 20 and 90 percent humidity. Many ideas were discussed that included direct injection of water into the gas flow, but due to the very low initial flow rates that were requested by the customer, support for a bubble humidifier increased. With a bubble humidifier, the gas enters a small water tank that is heated to control the temperature of the water. As the gas bubbles up from the bottom of the tank, it equalizes in temperature to the surrounding water and in the process absorbs a certain amount of moisture. By varying the temperature of the water, the humidity levels can be changed. The test stand will monitor the humidity of the gases downstream of the humidifiers and will adjust the temperature of the water to maintain a user entered value. The customer stated that they would be willing to run tests to collect empirical data for the humidifying process to determine a temperature vs. humidity curve for the actual system.

The humidifiers consist of a base which has channels allowing the gas to enter the tank, a polycarbonate or Pyrex tube that serves as the reservoir, and a polycarbonate top similar to the base. The top of the humidifier tank will contain holes for the cartridge heater, thermocouple, gas exit to the fuel cell, and a removable plug, so that water can be added to the tank. The humidifier is held together by a rod that protrudes through the tank and exerts a pressure on both the top and base of the tank, creating a strong seal.

[pic]

Flow System

[pic]

The current flow system consists of four main components; the flow regulators on the two tanks, the bubble humidifiers, the fuel cell, and the back pressure regulator. For our project, we are not providing the regulators to control the pressure from the gas tanks but we are considering the settings of those regulators when considering the flow control of the system. The two components that control the actual flow of the gases are the regulators at the tank that will provide a pressure that is slightly above the pressure that is desired for the test. At the opposite end of the system is located a back pressure regulator. The back pressure regulator is used to fine tune the pressure regulation within the system and also ensure flow through the system. The flow goes from the pressure regulators through a solenoid valve that can be shut if an error within the system is found to the humidifiers. Once the gas has bubbled through the humidifier, it is then carried through ¼ inch tubing to an “expansion valve”. The expansion valve is a chamber that the gas passes through that contains a small bread board with pressure, humidity, and temperature sensors attached to it. There are two useful features to using this expansion valve concept. First, it minimizes interference in the flow of the gas, by not inserting small probes directly into a 1/4” line. Second, if the sensors ever needed to be changed, the breadboard setup, can be easily interchanged outside of the test station.

[pic]

After passing through the expansion valve, the gas travels to a two position directional valve. This valve is installed along with two check valves to direct the gas to either the fuel cell or through a loop to the back pressure regulator. The loop is inserted so that pressure can be set prior to actually being directed to the fuel cell. This is to minimize possible damage to the fuel cell by having large pressure differentiation between the two gases. Once the pressure is set to the desired value that is measured from the sensors in the expansion valve, the directional valve is set to the run position and transfers the flow to the fuel cell. Once the gas flows through the fuel cell and out to the back pressure regulator, it is exhausted to a small water reservoir. The gas is exhausted to the water so that there is a visual indication of the gas flow as well as minimizing possible back flow into the system.

Electrical Components

The first step in the electrical system design was to determine which parts of the project could incorporate electrical components. Initially, it was determined that we could implement humidity, airflow, temperature, and pressure sensors, electrically. A data acquisition card was also determined to be needed to convert the sensor output signals into a format that could be read on a computer. It was eventually decided that resistive heating could be used for the heating of the humidity water tanks and then also in the water tank for fuel cell heating. Price was the major driving criteria for component selection. The components were researched extensively to find the cheapest parts for each function that would still satisfy the technical requirements of the project[1]. The best electrical sensors as determined by the operating capabilities and price are as listed in the following table.

|Sensors |

|Humidity |Honeywell HIH-3610-001 |

|Temperature |Honeywell 775-B-U-O |

|Pressure |Measurement Specialties 1451-100G-T |

|Airflow |Honeywell AWM2300V |

|Initial Sensor Selection |

Next, research was done to determine if resistive heating was feasible, and if so, what would work this project. Through research, it was determined that the best metal alloy for resistive heating was NiCr with a ratio of 70 parts Ni to 30 parts Cr. This ratio was determined to be the best because it provided the highest resistance and a high melting temperature. All of the voltages required to run the various components were now known, therefore the proper power source was researched. The initial solution for the power was to use a power converter to switch the standard 120VAC into 8V DC. Another power converter would then be used to switch from 8V DC to 5V DC for the sensors that require 5V. Next, the first overview schematic of the electrical schematic was created. It is shown below:

Initial Electrical System Schematic

At this point in the development process, the customer requirement for the gas pressure changed from 0-100 PSI to 0-30 PSI. Because of this, the pressure sensor was researched again, and the new solution became the Honeywell ASDX030A24R. Safety became a concern regarding using a capacitor in the H2 environment. Alternative electrical solutions were researched, but no satisfying solution was found. The mechanical team determined that the H2 concentration must be below 75% for there to be a risk of explosion. Therefore, it was determined to be acceptable to use a capacitive humidity sensor since the concentration of H2 will be nearly 100%. The customer specifications for the project changed again, and it was no longer necessary to measure the airflow of the gases. Without the airflow sensors, all of the sensors now used only 5V DC as their inputs. This allowed us to simplify our circuit by cutting out the 8V to 5V converter, and using a 120V AC to 5V DC converter instead. It was also indicated by the mechanical team that the project needed three heaters instead of one. One heater to directly heat the fuel cell, and one heater to heat each gas line’s humidifier tank. It was determined that three heaters require too much current for a 5V DC signal. It was decided that we will use the 120V AC directly from the wall to power the heaters. Each heater required nearly 50 feet of wire to keep the current at an acceptable level. To reduce the amount of wire needed, a resistor was added in series with the resistive wire. This reduces the amount of current the heater draws, however it also lowers the power rating of the heater. This version of the schematic looked like this:

[pic]

First Revision of Electrical System Schematic

There were some reservations about using a resistive coil to heat the fuel cell directly. This was because a resistive coil heater creates a large electromagnetic field around it, and the fuel cell is sensitive to such fields. Instead of directly heating the fuel cell, the resistive heater now heated water which is pumped through the fuel cell to heat it. It was determined that we needed underwater temperature sensors to keep the water from boiling. If the water boils, it can damage the pump and other internal components. Since all of the resistive heaters need to heat water, a new solution was found. Cartridge heaters are resistive heaters encased in a metal tube suitable for submersion. The feasibility of each component in the electrical system was then analyzed and the best components were officially determined. The mechanical team also indicated they would like an electrical sensor inside of the fuel cell compression assembly. These sensors were researched until suitable ones were found. Shortly thereafter, it was determined that a torque wrench would be used to compress the fuel cell assembly instead of an electrical pressure sensor. The following electrical components were then selected:

|Electrical Components |

|Humidity |Honeywell HIH-3610-001 |

|Air Temperature |National Semiconductor LM34 |

|Water Temperature |Honeywell TD4A Liquid Temp Sensor |

|Pressure |ASDX100G24R |

|Resistive Heating |Omega CIR-3016 |

|Second Sensor Selection |

The Honeywell TD4A temperature sensor does not have a linear output voltage which is used as the variable to determine the measurement, like all of the other sensors chosen. This sensor’s output variable is it’s current. Our DAQ card does not read currents. Therefore, a circuit was created to convert the output current into voltage. The following circuit achieves this:

Liquid temperature sensor, current to voltage conversion circuit

At this point, the project concepts selected were presented to the customer. The customer commented that they would like to have an auto-shutdown circuit incase of problems and a secondary heater shutoff incase of software hang. To address the issue of the secondary heater shutoff, an isolated electrical circuit was constructed that measures the temperature in the water tanks, and outputs a voltage high when a certain temperature is reached. The schematic for this circuit is the following:

Temperature comparator circuit

When the temperature of our water goes to 100 degrees Celsius, the output of this circuit will go high. This signal is then inverted and goes to the control signal input of a solid-state relay. Therefore, when the water boils, the control signal for this SSR will go low, causing the heater to lose power. One more SSR was added at the beginning of the entire electrical circuit. This will act as the automatic shutdown switch. When the software detects any dangerous readings, it will send a signal to this SSR which will cut off the power to the entire circuit.

Second Revision of Electrical System Schematic

The last change made in the Electrical system layout is the addition of the solenoids for the purpose of safety. These solenoids are closed by default and must be powered to allow gas flow into any part of the flow system. This change is shown through the Gas Control Valve block. Acceptable solenoids will be researched during the next quarter, but the approximate $100 which they cost, will be part of the overrun budget for the safety aspects of this project. The final schematic for the electrical system in this project is the following:

[pic]

Final Electrical System Schematic

The maximum current draws of each component were determined from their datasheets. The maximum current draw in this project is 2.57A. The calculation to determine the maximum time needed to heat the water tanks was also calculated. The mechanical team determined that 58200 Joules were needed to heat the largest tank from 20 to 80 degrees Celsius. The heating cartridges output 100W, therefore the maximum time needed to heat the tanks is 9.7 minutes.

The following Pugh charts were used in our electrical sensor selection:

Power Supply

|1 = much worse than baseline concept 2 = worse |Test Power |DAQ card power|One 120VAC | |

|than baseline 3 = same as baseline 4 = better |Supply | |source with | |

|than baseline 5= much better than baseline | | |converters | |

|Cost |1 |5 |3 | |

|Can it supply enough V and I |3 |0 |3 | |

|Connectivity with computers & Programmability |3 |3 |3 | |

|Ability to vary the V and I |4 |3 |3 | |

| | | | | |

|Mean Score |2.8 |2.8 |3.0 | |

| | | | | |

|Normalized Score |91.7% |91.7% |100.0% | |

Power Supply

|1 = much worse than baseline concept 2 =|Power IO - |Custom Made |Series 103 |DIN-A-MITE |  |

|worse than baseline 3 = same as baseline|3V50 |SSR |(Indeeco) |(Watlow) | |

|4 = better than baseline 5= much better | | | | | |

|than baseline | | | | | |

|Cost |3 |3 |2 |1 |  |

|Output Amperage range |3 |3 |3 |3 |  |

|Input Control |3 |3 |3 |3 |  |

|Dimensions |3 |1 |2 |1 |  |

|Reliability |3 |1 |3 |3 |  |

|Availability/Lead time |3 |1 |1 |3 |  |

|  |  |  |  |  |  |

|Mean Score |3.0 |2.2 |2.6 |2.2 |  |

| | | | | | |

|Normalized Score |100.0% |73.3% |86.7% |73.3% |  |

Data Acquisition

Due to the cost issues of the mechanical sensors, and the limitation of the project budget, most of the control and data monitoring will be performed using electrical sensors. The data acquisition system will be used to interface between the electronic devices and the software application of the fuel cell testing station.

The following table includes different kinds of sensors and the quantity that will be needed for building the fuel cell testing station. As a summary, there are a total of nine input signals and three output signals. The data acquisition system will have the capability to accept nine analog inputs, and to send three analog/digital output control signals.

|Temperature Sensors |

|Total |Usage |

|2 |One for each H2 and O2 gas line. |

|2 |One for each water tank. |

|1 |For mechanical assembly. |

|Pressure Sensors |

|Total |Usage |

|2 |One for each H2 and O2 gas line. |

|Humidity Sensors |

|Total |Usage |

|2 |One for each H2 and O2 gas line. |

|Heaters |

|Total |Usage |

|2 |One for each water tank. |

|1 |For mechanical assembly. |

NPRL, our customer already owns a copy of Labview, therefore Labview is the most feasible option to implement the software application. This also means that the data acquisition device should have Labview compatibility. The basic requirements of the data acquisition device are compatibility with Labview and the ability to provide at least nine inputs and three outputs.

DAQ Comparison:

[pic]

(1)ECQN DAQ

1) Pros:

• USB connection, support plug-and-play

• Compact, easy to hook up

Cons:

• Need two units to satisfy DAQ requirements.

• Need additional Labview driver

• Two device cost $ 458 + Shipping

[pic]

(2)Pico ADC-11

2) Pros:

• USB connection, supports plug-and-play

• Compact, easy to hook up

Cons:

• Need two units to satisfy DAQ requirements.

• Need additional Labview driver

• Two device cost $ 337.7 + Shipping

• Vender located in England.

[pic] + [pic]

(3)Measurement Computing – MiniLab + Drivers

3) Pros:

• USB connection, supports plug-and-play

• Compact, easy to hook up

• Two devices and the Labview driver cost $ 267 + Shipping

Cons:

• Need two units to satisfy DAQ requirements.

• Two device cost $ 290 + Shipping

• Need additional Labview driver, may have potential problems

[pic]

(4) National Instruments DAQ

(Winning Concept)

4) Pros:

• USB connection, supports plug-and-play

• Compact, easy to hook up

• Product of NI, supports Labview without additional driver

Cons:

• Need two units to satisfy DAQ requirements.

• Two device cost $ 290 + Shipping

The following is the Pugh’s method chart. Through the analysis, the USB data acquisition device from measurement computer is indicated as our best choice. It is the most cost efficient device. However, recent customer meetings have confirmed that the reliability of imitation National Instruments boards is not good, so the extra expense for the National Instruments NI USB-6008 is justified.

|  |USB-1208LS |NI USB-6008 |Mechanical |  |

| |(measurementco|( | |

| |) |.com) |Display | |

|Device Cost |3 |3 |1 |  |

|Compatible with LabView Software |2 |3 |1 |  |

|Be Able to Record Measurement to Computer |3 |3 |1 |  |

|Shipping Time |3 |3 |2 |  |

|Sampling Speed |3 |3 |3 |  |

|Number of I/O Channels Provided |3 |3 |1 |  |

|  | | | |  |

|Mean Score |2.8 |3.0 |1.5 | |

| | | | | |

|Normalized Score |94.4% |100.0% |50.0% | |

Software Application

The software development will work in stages. Before the completion of each development stage verification and validation should be performed. According to the data acquisition system needs-assessment process, the fuel cell testing station needs a total of nine input signals and three output signals. The detailed requirements that translated to the software requirements are:

Required Testing Parameter (Provided by User):

• Desired H2 and O2 gas humidity.

• Desired H2 and O2 gas pressure.

• Desired PEM temperature.

• Desired PEM pressure.

• Maximum allowed temperature for each H2 and O2 water tank.

• Maximum allowed PEM pressure (mechanical press pressure).

• Testing duration.

• Output file type and name.

Temperature:

• Display the temperature of the H2 and O2 gas line.

• Display the temperature of the H2 and O2 water tank.

• Use the measured temperature of the H2 and O2 water tank to control the operation of the two heaters in each of the H2 and O2 water tank.

• Display the temperature of PEM water tank.

• Use the measured temperature of PEM water tank to control the operation of the heater in the PEM water tank.

Humidity:

• Display the humidity of the H2 and O2 gas line.

Pressure:

• Display the pressure of the H2 and O2 gas line.

The software application for the fuel cell testing station contains two parts. One is the system operation control on the back end; the other part is the user interface front end. When designing the system control internally, the most important thing is to use the flexibility of software to integrate in the safety assurance capability. When designing the user interface, understanding what the customer wants to do with the software, and what are the operations that they will perform frequently will help guide the software design. Software designed based upon these considerations will make the most frequent tasks the most convenient.

According to the detailed software requirement, the following are the software design flowcharts.

[pic]

At the start of the software program, it will test all the temperature sensors of the fuel cell testing station. If any of these temperature sensors is operating abnormally, software will show the corresponding error message and terminate the program to ensure safety. The logic behind this, is that we cannot guarantee the heater will only heat the water to the desired temperature. If any of the heaters keep operating without any control, this could lead to a hazardous environment where H2 was being heated past acceptable levels. In the fuel cell heating tank, if the water began boiling due to overheating, the pump would begin to operate improperly and would likely cause irreparable damage. Software should detect the situation, and prevent such dangerous situations from occurring.

[pic]

The above is the testing environments set up logic flowchart. Except monitoring sensors and controlling the heaters, safety mechanisms are also implemented throughout the process. For example, user will be asked about the maximum temperature that each water tank can be heated to, and the software will continue checking the water temperature. If the temperature exceeds the maximum value, then a warning message will be shown. If a warning message is generated, the software will continue to check the water temperature. If the temperature continues on hitting improper values, and reaches the dangerous level, the software will automatically shut off the heater to ensure safety. A similar mechanism is also used to check the H2 and O2 gas pressure.

Budget

|Item |Quant |Cost |Total Cost |Name |

|Electronics |  |  |  |  |

|DAQ |2 |150 |300 |National Instruments NI-USB 6008 |

|Temperature Sensor |2 |10.08 |20.16 |Honeywell 775-B-U-0 |

|Humidity Sensor |2 |21.42 |42.84 |Honeywell HIH-4000-001 |

|Pressure Sensor |2 |20 |40 |Honeywell Microstructure Pressure Sensor ASDX030G24RDO |

|Solid State Relay |3 |22.94 |68.82 |HDA-3V25    |

|Breadboard |2 |5 |10 |  |

|Current Source |1 |  |NPRL |Keithley 220, .5pA to 101mA output |

|Voltage Measure Device |1 |  |NPRL |Keithley 195A |

|GPIB Interface for Current/Voltage |1 |  |NPRL |  |

|Computer with Labview |1 |  |NPRL | |

|  |  |Subtotal: |481.82 |  |

|Casing/Safety |  |  |  |  |

|Sheet Metal |4 |24.4 |97.6 |1/16"x24"x24" Al 1100 - 88685K18 |

|Lexan |4 |33.54 |134.16 |McMaster-Carr 1/4"x24"x24" (12ft-lb/in) |

|Exhaust Fan |1 |10.98 |10.98 |MASSCOOL FD12025S1L3/4 120mm Sleeve Case Cooling Fan |

|Exhaust Tube |1 |1 |20 |  |

|  |  |Subtotal: |262.74 |  |

|Humidity |  |  |  |  |

|Tank |2 |20 |40 |2" Inner Diameter Polycarbonate |

|Connections |4 |5 |20 |  |

|Cartridge Heater |2 |31 |62 |Incoloy Sheath Immersion Heating Cartridge |

|  |  |Subtotal: |122 |  |

|Flow System |  |  |  |  |

|Oxygen Pressure Regulator |1 |183.14 |NPRL |McMaster-Carr Two Stage Pressure Regulator 5/8"-18RHF |

|Hydrogen Pressure Regulator |1 |  |NPRL |Oxweld Heavy Duty Trimline R-77 Regulator |

|Back Pressure Regulator |2 |89.57 |179.14 |1/4" Aluminum adjustable pressure maintaining relief |

| | | | |valves - 4783K51 |

|Tubing |1 |10 |10 |50 ft at |

|Tube Fittings |28 |5 |140 |  |

|Quick Connect Tube Fittings |8 |7 |56 |  |

|Exhaust Bubble Jar |2 |9.81 |19.62 |Oil Cup Cylinder 2"Dia, 1-7/8"H - 1176K14 |

|Expansion Valve |2 |20 |40 |  |

|  |  |Subtotal: |444.76 |  |

|FC Heating |  |  |  |  |

|Water Tank |1 |30 |30 |  |

|Cartridge Heater |1 |31 |31 |Incoloy Sheath Immersion Heating Cartridge |

|Water Hose |16 |0.75 |12 |Tubing 3/8" - 8585K11 |

|Reducing Union |2 |5.56 |11.12 |1/2" to 3/8" Pump to Tube Reducing Union - 44705K272 |

|Tube Fitting Tee |1 |10.4 |10.4 |3/8" Pipe to 3/8" Compression Tube Fitting Tee - 2227K42 |

|Reservoir Tube |6 |3.4 |20.4 |3/8" Pipe to 3/8" Compression Tube Fitting - 2227K12 |

|Tube to Fitting |8 |0.23 |1.84 |3/8" Compression Fitting Nut - 50915K121 |

|Water Pump |1 |33.59 |33.59 |Pentair Aquatics Rainbow Lifeguard Quiet-One 1200 |

|  |  |Subtotal: |150.35 |  |

|PEM Mechanical Assembly |  |  |  |  |

|1' x 1' x .5" Aluminum Plate |1 |51.85 |51.85 |1'x1'x.5" Aluminum Plate - 9246K33 |

|3/8" Aluminum Rod |1 |27.34 |27.34 |Ground, 36" Length - 9062K283 |

|Precision Modified Nut |1 |18.4 |18.4 |1/2"-10 Internal Thread - 6350K41 |

|Power Screw |1 |9.27 |9.27 |1/2"-10 Power Screw - 93420A110 |

|Bronze Plain Bearing 3/8" |2 |0.43 |0.86 |3/8" ID 1/2" OD 1/2" Thick - 2868T6 |

|Bronze Thrust Bearing |4 |0.65 |2.6 |3/8" ID 3/4" OD 1/8" Thick - 2879T3 |

|Bronze Plain Bearing 1/4" |2 |0.56 |1.12 |1/4" ID 3/8" OD 1/2" Thick - 2868T3 |

| | |Subtotal: |111.44 | |

| | | | | |

| | |Total: |1573.11 | |

The budget was laid out as one of the top priorities of the project from early on. The range given to us was a $1000 to $2000, with a hard cap at $3000. It was quickly apparent that we would not be under budget on this project. There are many sensors and mechanical systems to be accounted for. These systems are laid out in the budget list above. The largest portions of the project as can be seen are the Electrical Sensors and Flow System, each of which is roughly $450 or almost 1/3 of our current budget.

After finding out the budget available to the team, a price quote was requested from TVN Systems for their RU-2100 model which is in the Mechanical Engineering Fuel Cell lab. In the early goings of this project it was expressed by the customer, that if we could duplicate that device, our project would be a success. The $29,995 price tag on the commercial model, was however out of the range the NPRL was willing to spend.

At the midway point of this project we are right on budget, due to extensive research into electrical components that will meet our testing needs for the absolute best bargain. All mechanical systems are designed to require the least number of parts with the cheapest grade material that will still perform the function necessary. Unbeknownst to our team until late in the Senior Design I process, we actually managed to design a large portion of our test station, in the same manner as the TVN Systems model. Our method of back pressure regulation is the method used to control the flow in the RU-2100, however they control the flow rates, where as NPRL has not expressed an interest in knowing the flow rates. At the least they have not expressed the interest in paying over $200 extra for the two flow sensors required to find a flow rate in the system. We are also providing a theoretically more accurate heating system, and a device to clamp together a PEM assembly, which is not included in the RU-2100.

The goal of this project is to produce a working test station by the end of Senior Design II, if this is successful, it leads to very interesting possibilities. It has been expressed to the team, that the commercial desire for such a test station is very high, because there are few in existence. Looking at the capabilities and accuracies of the proposed system, they are comparable to the RU-2100 model. The price on the other hand is a factor of 10 better. Even with the additional features necessary to match the RU-2100, our cost will be negligible. In addition, all the man-hours required for the design of such a system, will have been completed at no cost during the Senior Design process. This leaves a strong possibility that assuming our prototype functions to specifications further production for commercial firms would be possible.

Conclusion

There have been several changes of scope to this project over the course of the quarter, however, the group has handled change well, and has a viable solution to testing the NPRL experimental gas diffusion layer. The work break down structure as laid out above defines our duties as individuals in the team well. This quarter has made everyone truly look at the customer needs and the project goals for this project, but next quarter must have an emphasis on the assembly and testing of this test station. The key to Senior Design II will be the early ordering of crucial parts, especially the electrical sensors, possibly as early as this finals week. The technical skills of this team will easily allow the project to be completed, assuming we have the time necessary with the parts in our hands, so we can begin construction and debugging.

-----------------------

[1] Note: The datasheets for every electrical component named in this report can be found in the team folder.

-----------------------

➢ 8 Single-Ended or 4 Differential Analog Inputs

➢ 12-bit Resolution

➢ Two 10-bit Analog Outputs

➢ One 32-bit External Event Counter

➢ 16 Digital I/O Lines

➢ External Trigger Input

➢ Sample Rate 10KS/s

[pic]

[pic]

Power Converter

5V DC

120V AC

8V DC

To/From Software

Data Acquisition Card

Heat source

5V sensors

8V sensors

Power Converter

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download