Design Document Final Semester 2 - May15-30
Contents
1. 6 3 ACRONYMS AC Alternating Current CAD Computer Aided Design DC Direct Current EMF ElectroMotive Force LED Light emitting Diode MO Magneto Optic MOSFET Metal Oxide Semiconductor Field Effect Transistor PCB Printed Circuit Board TMS Transcranial Magnetic Simulation Vp p Voltage peak to peak 6 4 UNIT VARIABLES B Magnetic field of a coil tesla conversion 1 tesla 10 000 gauss u Permeability of free space 47 x 1077 henries meter N Number of turn of a coil dimensionless Current through a coil amperes R Radius of a coil meters L Inductance of a coil henries l Length of a coil meters May 15 30 EI 7 APPENDIX C FIRST DESIGN The original design of our circuit was created in EaglePCB and then fabricated on the Protomat S62 We chose this technique as it allowed us to make quick changes to the PCB and fabricate a new board in a timely manner Shown below in Figure 29 are examples of some boards we created Ultimately these boards did not perform to the satisfaction we wanted Between issues with soldering weak traces very small package sizes etc and our desire to have a clean looking product we ultimately re designed our board and had it professionally fabricated Figure 29 Examples of Boards As our knowledge of how our circuit needed to operate improved we learned of various things we could improve in our circuit for our new design We wanted to make the following changes Inc2. TOP LAYOUT Oscilloscope Input Waveform Generator Input 500 Resistor Red LED 9310 Resistor 11V Zener Diode 12V Zener Diode Current Sense 100V Diode Resistor Coil Inputs MOSFET 20 Resistor 100 pF Capacitor 17V Zener Diode 100 pF Capacitor 1 3kQ Resistor 1 uF Capacitor 01 uF Capacitor j g 16V Zener Diode 1 43kQ Resistor Red LED DC Barrel Jack Green LED Figure 26 Partitioned Circuit Board May 15 30 5 2 SETUP Along with the PCB you will need the following equipment to correctly set up the circuit e Coil e DC Power Cable e Oscilloscope and Probe e SMA SMA connectors x2 with SMA BNC Adaptors x2 Figure 27 Coil Examples SMA BNC emp SMA BNC SMA SMA Figure 28 Cable Connector Examples 5 2 1 CONNECTING CABLES To connect the PCB to the oscilloscope DC supplv and waveform generator correctiv 1 Make sure DC supplv and waveform generator are OFF 2 Insert and screw one lead of vour coil in the left port of vour wire to board terminal and insert and screw the other lead of the coil in the right port Mav 15 30 a 3 Connect 1 BNC SMA adaptor on each of the SMA SMA connectors effectivelv creating 2 BNC SMA connectors 4 Connect 1 BNC SMA connector from the Current_Sense SMA on the PCB to the oscilloscope 5 Connect 1 BNC SMA connector from the AC_Input SMA on the PCB to the waveform generator 6 Connect the DC Barrel Jack of the PCB to a DC su
3. coil and a single coil configuration When constructing coils we use a 6mm diameter 3 mm radius coil The length of a five turn coil is close to 1 mm in length Below are the coil calculations for each of type of coil Ultimately speed of our pulse width is very important and we decided to choose the single coil design 1 3 7 SINGLE COIL CALCULATIONS We calculated the magnetic field for a single coil using the equation below May 15 30 ES uNI 4r x 1077 5 48 41 VE 4R Ji 4x10 Coil inductance for the single coil is found using the following equation N nR An x 1077 5 m 3 x 1079 7 VZ 4R2 J 1x 10 3 4 3 x 10 32 1 3 8 HELMHOLTZ COIL CALCULATIONS 0 050006 Teslas 500 05 gauss 146nH The magnetic field for a Helmholtz coil is found by using equation below p 0 7155 N7 _ 4m x 10 7 0 7155 5 33 37 0 T d R 3x 10 3 0 050006 Teslas 500 06 gauss Coil inductance for Helmholtz coil is found using the equation below L 2u 0 7155 N2nR 2 4r x 1077 0 7155 57 1 3 x 1073 423 7nH 2 DETAILED DESCRIPTION IMPLEMENTATION DETAILS 2 1 SYSTEM DESCRIPTION Figure 5 shows a block diagram of the concept of our design This outlines the process of taking a pulse generator and sending a one microsecond pulse to a MOSFET The MOSFET then switches on causing a DC current to be drawn through the coil In accordance with Faraday s law the current driven through the coil will generate a magneti
4. enough because their sample rate is lower than the required 2 MHz we would need The system will send out an optical signal that will split at a coupler and then interfere with itself resulting in either constructive or destructive interference which will be measured at the output The interference depends on the state of magnetization of the magneto optic MO material In one direction we would measure no output and when it is magnetized the other way we would see an output Laser 1550nm MO L Coupler E Material f U e Magnetic ower Sensor Field Driver Oscilloscope Figure 18 Sagnac Interferometer May 15 30 ES As can be seen in Figure 18 above the Magnetic Field Driver is located around the MO Material This is where our circuit would be The MO material is housed in a sheath between two optical connectors that are coupled together as shown in Figure 19 below Our testing would measure the optical signal s power over a 360 range and then measure it again after pulsing our magnetic field to change the polarization of the MO material Figure 19 Sagnac Interferometer Configuration 3 6 2 INITIAL DATA The important data for us to measure was that we could generate our pulse in 1 us and also generate a field that was at least 500 G Using a 15 V DC input 1 us pulse width input 2 Vp p and coil C from section 1 3 6 we were able to obtain the following data shown in Figure 20 below The results show that our pulse w
5. in a short period of time 1 us This carried the inherent risk of overloading our circuit components and rendering them useless Thus all of our components were specifically chosen to be able to withstand a high current Even though all of the components were chosen to be able to withstand a short burst of high amperage care was still executed during testing At one point during testing we were changing the pulse width to test the effects on our circuit output When the pulse width was increased without increasing the period the MOSFET essentially remained on and our coil was overloaded due to continuous current being driven through it After this incident whenever the pulse width was changed the period of the pulse was double checked 3 7 4 COLD SOLDER During the testing of the first design a cold solder joint was discovered Cold solder is where solder does not melt completely at the joint therefore leaving a layer of oxidation in between the component and the board With that layer in between the component and board the component was not connect to the board as assumed To repair this issue the component was removed and re soldered on 3 7 5 DC BARREL JACK SCHEMATIC While testing the second design we discovered the DC barrel jack functioning contrary to how we expected After looking at the datasheet we determined pin A was our ground pin and pin B May 15 30 ES was a wedge inside that indicates whether something is plugged in or no
6. in this simulation where one inductor coil has twice the inductance as the other It is important to note that the inductances are still very small but have a large impact on our circuit RS c1 c2 R8 Ce c4 c5 2 220uF OtuF mu 2 T za omu Ap K eee 146nh oo Y 282nh 1N4500 D 1N4500 D DRAIN SOURCE GATE PSMNORS 30YLD v3 1svde 1svdeal DRAIN SOURCE GATE R7 PSMNORS 30YLD 05 v1 0 V120 v4 V2 5 V2 5 R6 TD 0 TD 0 49 9 TR 0 TR 0 TF 0 TF 0 PW 1us PW 1us PER Ims PER 1ms D 0 Figure 12 Inductance Effects on MOSFET As shown in figure 13 the inductance of the left circuit is 146nH and the inductance on the right is 292nH The circuit with the smaller inductance is represented by the purple line in Figure 13 below and the green line is the larger inductance May 15 30 46A 20A 1us o I L1 I L2 Figure 13 Pulse Width Measurements As shown in figure 13 above the smaller inductance purple line had a faster rise time then the larger inductance This led us to decide that we needed a coil with a small inductance to minimize our rise time and maximize our magnetic field When simulating this circuit initiallv we saw that the fall time was extremelv long and postulated that the long rise time was due to the fact that our coil was reenergizing itself as it discharged To combat this we added a 2 ohm resistor to our feedback protection circuit Vou can see it as
7. nH legend hi h2 Current Inductance tend HELMHOLTZ CODE 9 2 1 MAGNETIC FIELD CALCULATIONS May 15 30 cyfiles gt megsharp gt Documents MATLAB gt EE492 yfiles megsharp Documents MATLAB EE492 wodInn obt NNNNNNNNRPP RPP PRP PB BB A oO oOwm A eum e Who IdCalculationHelmholtzCoil491 m clear all close all Hmiz Hon ow ki ki kd lasta BfieldCalculationHelmholtzCoil491 m CurrentCalculationHelmholtzCoil491 m ILvs_N_table_Helmholz m B_T 4 pi 10 7 0 7155 N I R 0 001 Bg BT 1004 u N I B R UNITS B gt Tesla T gt V s A m u gt H m gt Wb A m gt V s A m N gt dimensionless I gt Amperes A R gt radius m V s V s A m A V s m V s m 2 m m m 2 9 2 2 CURRENT CALCULATIONS input How many turns are in the coil input What is the radius of the coil in mm input What is the current of the coil in Amps 3 Then we found that 1 Gauss 0 0001 Tesla N gt of turns in the coil dimensionless R gt radius of the coil meters 3 I gt current of the coil amperes k Equation from MFG Design paper for a Helmholtz coil Converting magnetic field units from Tesla to gauss disp The magnetic field generated with those parameters for a Helmholtz coil is num2str B g Gauss jJ To verifv the equation used above we did a dimensional an
8. 20 Resistor 100 uF Capacitor 17V Zener Diode 100 uF Capacitor 1 3k0 Resistor 1 uF Capacitor 01 uF Capacitor Ore 16V Zener Diode 1 43kQ Resistor Red LED DC Barrel Jack Green LED Figure 16 PCB Layout DOSO 0 KE A l U M Figure 17 Populated Circuit Mav 15 30 3 5 SYSTEM FEATURES In addition to the required functions of the PCB many features have been added to make it easy to use and understand in comparison to our initial designs A DC Barrel Jack is included to allow the user to plug the DC voltage into the PCB Board Edge SMA Connectors simplify the process of hooking the pulse input and oscilloscope They also keep the wires out of the way of other components of the circuit The green Wire to Board Connector is used as the input for the coil This component allows for coils to be switched in and out Throughout the board you will find LEDs The green LED indicates when the DC power supply is working Two red LEDs indicate when the circuit may be in harm due to the voltage either DC or AC being set higher than the intended value 3 6 DATA ANALYSIS 3 6 1 INITIAL TEST SETUP To perform testing using the client s Sagnac interferometer setup we needed to gain a better understanding of what the Sagnac interferometer does Shown below in Figure 18 is the Sagnac interferometer This setup is used to detect the magnetic field pulse Conventional magnetic field detectors are not able to pick up the pulse fast
9. 3 ENERGY DISSIPATION A 20 resistor in series with the diode is placed in parallel with the wire to board connector This is because without the resistance the current would continue to loop between the coil and May 15 30 al diode because there is no place for the energy to dissipate OrCAD makes all wire resistances ideal zero thus we may not need this 2Q resistor in our physical circuit The wires and connections may cause enough resistance for this energy to dissipate elsewhere Our design also contains a current sense resistor with a value of 0 050 Due to the difficulty of measuring the magnetic field our coil can output the current sense resistor will allow us to measure the drain current which should theoretically match the current going through the coil For example if we are expecting to use 20 25A we should expect to see a 1 00 1 25V drop across the resistor thus changing our gate to source voltage from 5V to 3 75 4 00V It is something we need to account for and ideally the resistor would not be a part of our circuit It is first and foremost for testing purposes 1 3 4 SWITCHING DEVICE Through preliminary simulation we were able to determine that our MOSFET was going to need to handle at least 65 amps of pulsing current Most importantly this MOSFET meets a number of design specifications Drain to Source Voltage VDS must be able to handle 15V or more Current channel must be able to handle the current load through t
10. As explained in previous sections the coil provides the magnetic field and the MOSFET is used as a switch The members of our team split into two subset teams coil and MOSFET circuit teams The coil team began research on the type material and inductance of various coils The team focused on achieving low inductance in order to have a quick rise and fall time Also in order to meet our design requirement of reaching 500 gauss the current number of turns and wire specifications needed to be evaluated The MOSFET team created the circuit schematic in order to handle fast switching and high current Optimization of circuit and addition of components that complemented the MOSFET and coil were next in the process Simulation was done via OrCAD to verify appropriate rise and fall times If we did not reach appropriate rise and fall times the inductance of the coil needed to be lowered Once rise and fall times were met in order to get our one microsecond pulse the PCB layout was created Fabrication and population of board were next Once we had a functional board the testing procedure began 3 2 TEST PROCESS Testing was a key process in the second semester Once the board was populated a continuity check of the circuit was conducted as shown in figure 8 When components didn t pass the check they were re soldered The one microsecond pulse width was the next obstacle to overcome The rise and fall times needed to be quick in order to achieve t
11. For the final design a GERBER file containing all the information needed to fabricate a board was sent to OSHPark for a professionally fabricated board For the professionally fabricated boards on the final design the method of reflow soldering using a reflow oven was May 15 30 ES implemented Using the reflow method is faster and more accurate especiallv for smaller components in comparison to soldering bv hand Mav 15 30 Le 3 RESULTS The overall design process proved to be very complex however the test plan demonstrated the most complexities We attained results that met all requirements that will be explained in later sections but first we will look at how we got to that point Starting with a simple idea this project evolved into a finished fabricated project based on rigorous scheduling extreme thoroughness and cooperative efforts Design Process 3 1 DESIGN PROCESS Test circuit Measure pulse Circuit soldered width correctly continuity Measure Test Process Pulse width voltage across Evaluate met current sense current resistor Calculated Fiber optic magnetic field eg Test sting strength met ch Figure 7 Design amp Test Process May 15 30 Le The design process and test process were integrated throughout the entire project We started our design process by quickly realizing that the coil and MOSFET were going to be the two most important elements of the magnetic pulse generator
12. MAV 15 30 Final Document Fast Compact High Strength Magnetic Pulse Generator MAY 15 30 ADAM KAAS BRITTANY DUFFY GREGORY FONTANA MEGAN SHARP BRANDON DIXON MEIYONG HIMMTANN ALAIN NDOUTOUME 1 CONTENTS 1 System Level Desig Zeegbgege e eegeehtt Geier inian iaio ibni pis i Siiani 4 LI S mmary EE i iii ia e u seb e eau 4 1 2 Project e UE Ee 4 1 2 1 Flin tiotal PRS UIT ES a i a a 4 1 2 2 Non Functional Reouirements EE 4 13 Functional De ompositieti ena asaadeicinmcemaien 5 1 3 1 Power SUP OI Y ika ke a AENA SOAN AKRE EEES 5 1 3 2 Energy StOa i L I 5 1 3 3 Energy DISSIPATION 1 e cbsetdee gege e ode eeh SedeeegdEe ia codusndendsanshiadiestsendecnesneedentdene 5 134 Switching TE 6 SC Ce E H 1 3 6 Single Coil vs Reimdbwaltseegk ege tege ege di ett Eed 9 1 3 7 Singl Coil Calculations iii ta kerkeifegua a Sii kat G tr wi a 9 1 3 8 Helmholtz Coil Calculations ise tii tai ius jesa name 10 2 Detailed Description Implementation Details iii i i EA i ri 10 2 1 System Description 10 2 2 Wlser ne Ateni xtiavnisetesmiicoesiecniestoumiaita daha husilanche soins Sioa E 11 2 3 Fabrication Methods EE 11 3 RES nikri e a nnn EE sie EAE a daa pea a E Eat 13 3 1 DESEM ee 13 3 20 TESEPFO ESS ii rii ia e aa Sendak rd a a a d a 14 aa SNM AOU EE 15 CN DE ae eu 15 3 3 2 Simulation Results ssis tege 16 BA HardWare eebe 19 3 4 1 Printed Circuit BOANO i kisi ki gnien tkkedat tasakesi Gie adian naaie 19 3 5 System Fea
13. R5 in the left circuit on Figure 14 below RS i 2 bad xl ae Lad al e Dm Si 146nh D2 E 292nh 1N4500 a 1N4500 Ja 15vda PSMNOR9S 30VLD S V1 0 V1 0 Ka v2 5 V2 5 TD 0 TD 0 TR 0 TR 0 TF 0 TF 0 PW tus PW tus PER 1ms PER 1ms D 0 Figure 14 Testing with Back EMF Protection Mav 15 30 The addition of the 2 ohm resistor decreased the fall time substantiallv as seen in figure 15 below Figure 15 Resistor Effects on Fall Time The circuit containing the resistor and diode in parallel with the coil is seen as the magenta line while a circuit with only a diode in parallel is seen as blue The circuit containing the resistor in parallel with the coil de energizes in a fraction of the time that circuit without does 3 4 HARDWARE 3 4 1 PRINTED CIRCUIT BOARD Figure 16 depicts the lavout of the PCB using CadSoft EAGLE Schematic Editor Appendix C illustrates previous designs from last semester that were altered into the lavout vou see before you here The dimensions of this board meet our specifications totaling 2 0 L x 1 5 W As can be seen each component on the board is labeled to make it user friendly See Appendix A for the user guide for more information on the circuit board May 15 30 e Waveform Oscilloscope Input Generator Input 500 Resistor Red LED 9310 Resistor 11V Zener Diode 12V Zener Diode Current Sense 100V Diode Resistor Coil Inputs MOSFET
14. alvsis Mav 15 30 f A gt Documents MATLAB b EE492 E Editor iastate edu cyfiles megsharp Documents MATLAB EE492 I_vs_N_table_Helmholtz m i BfieldCalculationHelmholtzCoil491 m CurrentCalculationHelmholtzCoil4 l m Lvs N table Helmholtz m E r input What radius do you want to graph in mm 2 SHS N 1 1 20 E I zeros 1 20 GH VE L zeros 1 20 6 7 for j 1 num rad De for i 1 1 20 CIE I i 0 05 r 001 N i 4 pi 10 7 0 7155 IE L i 1000 2 4 pi 10 7 0 7155 N i 2 pi r 0 001 ai end 12 figure j tal hold on 14 hAx hi h2 plotvv N I N L 15 str sprintf Turns vs Current Single Short Coil R tfmm r SEN d title Helmholtz Coil 500 Gauss Parameters a a xlabel of Turns txlim 0 50 Lips ylabel hAx 1 Current Amps tylim 0 100 19 vlabel hAx 2 Inductance nH 20 legend hi h2 Current Inductance 2r en Mav 15 30
15. ange in the power of the optical signal would be close to zero 3 6 4 FINAL DATA We were able to discover that when we set the function generator to create a pulsing waveform with an amplitude of 3 Vp p we observed saturation in the MO material as shown in Figure 22 below The vellow line represents the voltage measured at our CSR while the blue line is measuring the change in power of the optical signal As stated previously we observed this as the saturation point because of the behavior of the optical signal power showing no change Max Std Dev miu width 3 774jus Low signal amplitude B i 1 25GS s l F 44 0mv JR 10k points j E set Horiz na T i FE EE Mode i waveform XY Display 5 l average Unat GK EE Figure 22 Saturation Observed 3 6 5 THEORETICAL VS ACTUAL The results we obtained in our final testing demonstrated some abnormalities we did not expect to encounter Based on the experimental results our circuit saturated the magneto optic MO material implving that a magnetic field strength of 225 G was obtained Based on the parameters of our coil length radius number of turns the voltage measured across the current sense resistor and the magnetic permeabilitv we obtain a theoretical value of Mav 15 30 approximatelv 112 G There are manv factors that we knew could affect our theoretical calculations such as The magnetic permeability may be larger due to us assuming that u 1 Th
16. c field The current sense resistor CSR is in place strictly for measurement purposes The current going through the CSR is nearly equivalent to the current going through the coil From there we are able to see the voltage output via the oscilloscope The voltage is measured across the 0 050 CSR as well as the pulse width Utilizing the output voltage and the resistance 0 05 Q we can calculate the current Once we have the current value we use the can use the magnetic field strength equation to find the strength of the magnetic field generated May 15 30 EW Current driven through a coil DC Source l aon will generate a magnetic field Current Sense Pulse Generator Oscilloscope Resistor Figure 5 System Description 2 2 USER INTERFACE The circuit can be used in a variety of setups depending on the coil Shown below in Figure 6 is an example of how our circuit was used in our client s Sagnac interferometer setup The coil is placed over a small tube that has a magneto optic material located inside that can have its polarization changed depending on the various parameters for the circuit For more information view the User Manual in Appendix A Figure 6 Interferometer Setup 2 3 FABRICATION METHODS In creating the first version of prototypes a PCB prototyping milling machine called the ProtoMat S62 was used to cut out boards for quick prototyping Soldering done on the first versions of prototypes was done by hand
17. ch and simulations we found that having a low inductance was also important The lower the inductance of the coil the faster the rise and fall time of our pulse width would be In order to achieve low inductance in the coil we looked at what parameters UNI ant and L uN TR ART and found that the number of turn of the coil was a parameter that impacted both the length number of turns etc we could change We reviewed equations B current and the inductance the two most important parameters for a strong field and fast pulse We decided to create a MATLAB script that can be found in Appendix E to plot these parameters as shown below in Figure 3 May 15 30 ta EI Figure 1 5 8 28 File Edit View Insert Tools Desktop Window Help DOSGMs k AAVOPEL a a 0 Single Coil 500 Gauss Parameters 400 2000 Current Inductance 300 1500 T e 20 1000 2 E E o o 5 E o 100 0 2 4 6 8 10 12 14 16 18 20 f of Turns Figure 3 Single Coil Parameters at 500 Gauss Current and Inductance vs Number of Turns of the Coil As shown as the numbers of turns increases the current decreases and the inductance increases From this we knew that we wanted a fairlv small number of turns in our coil Fora single coil somewhere around 5 turns would give us a high enough current to produce 500 gauss and not frv our components while giving us a small enough inductance to meet our pulse width speed requireme
18. e coil is not perfect small gaps between turns not perfectly aligned etc The effects of the sheath housing the MO material Sensitivity of the MO material Human errors in measurements This is an issue we are still working through currently At this moment although our theoretical calculations don t appear to demonstrate functionality our circuit is performing to the satisfaction of our client 3 7 IMPLEMENTATION ISSUES CHALLENGES 3 7 1 SOLDERING SMALL COMPONENTS Packaging sizes of the components ranged from a 0 01uF capacitor with the dimensions of 1 0 mm x 0 5 mm 0402 to a general purpose diode with the dimensions of 7 95 x 5 90 x 2 25 mm DO 214AB SMC With one of the requirements of the design being the size of the device our first design measured 1 49 x 1 3 At that size and the sizes of the components soldering of the components onto the board was difficult Figure 23 Small Component Example 3 7 2 TRACES LIFTING OFF While soldering components onto the first board created from the ProtomatS62 the copper board used had traces that would lift off when removing the soldering iron after attempts to solder on a component This issue was resolved when ordering a professionally fabricated board from OSHPARK May 15 30 A Figure 24 Effects of Repeated Soldering 3 7 3 HIGH CURRENT RISK OF BURNING COMPONENTS The design of our circuit is focused on driving a large current about 60 amps though a coil
19. e use of two power supplies One DC power supply and one pulse generator The DC power supply will provide the current for energizing the coil We will have the power supply to the coil set at 15 volts This will power the capacitors that drive the current through the coil Our second power supply is a pulse generator This power supply acts as a trigger for our MOSFET It will send a five volt pulse every one microsecond 1 3 2 ENERGY STORAGE Capacitors near the DC voltage input hold a charge and effectively reduce the time it takes for the current to pass through the coil Using our present calculations for the current and voltage we have decided to use four capacitors of values 01uF 1uF and two 100uF The equations below gave us a minimum base value we could use for the total capacitance This equation assumes that we have 25 Amps of current in our inductor and a pulse width of one microsecond Current x Pulse Width SE Capacitance 25 Coulomb x 1 x 10 seconds second 167210 fFarads 15 Volts For a circuit like ours it is common practice to take this minimum value and multiply it by 100 or more Our total capacitance used in our circuit is 200 11uF which is roughly 120 times larger than our base value The 1uF and 100uF capacitors we are using are tantalum capacitors because of their lower internal resistance while the 0 01uF capacitor was ceramic due to the cost and availability of a tantalum capacitor that small 1 3
20. he coil Surface mount component Fast switching N channel operates using a positive input voltage and faster switching Suited for high switching frequencies resulting in a soft recovery Low parasitic inductance and resistance VDS rated up to 30 V Pulsed peak drain current rated up to 1888 A EMF voltage occurs when the waveform generator goes from its upper voltage to lower voltage In EMF voltage the change in current over the change in time becomes negative The inductance doesn t change hence giving a negative voltage across the coil Because the change in time will be very small the magnitude of the negative voltage will be very large Without a diode this large negative voltage will reach the MOSFET and potentially ruin the MOSFET We have chosen a large diode in parallel to the coil that prevents back EMF into the MOSFET Specifications include Fast recovery time High reverse breakdown voltage May 15 30 ES Surface mount component 1 3 5 COIL The coil is a necessary part of the circuit as explained by Faraday s Law whereby a current passing through a coil will generate a magnetic field The current is also directly proportional to the magnetic field so to increase the magnetic field to 500 gauss we wanted our circuit to be able to handle a higher currents The magnetic field is generated at the center of the coil as shown in Figure 2 Figure 2 Magnetic Field Generation During our resear
21. his requirement When a one microsecond pulse was generated an oscilloscope measured the voltage crossing the current sense resistor As previously discussed the current through Figure 8 Continuity Check the current sense resistor is the same as the current through the coil This current value was used in calculating the magnetic field Towards the end of testing we realized just how difficult it was to be able to prove through testing not just theoretically that specifications were reached due to measuring instruments not able to read the magnetic field of such a short pulse By testing using our client s Sagnac Interferometer setup we were able to confirm proof of specifications met e Initial testing plan and procedure e How we tested and treaded lightly at first e Initial setbacks e Cable issues e Switching from manual to automated testing May 15 30 mi 3 3 SIMULATION 3 3 1 CIRCUIT SCHEMATIC Shown below is the schematic of the magnetic pulse generating circuit As you can see by the organization of components this schematic closely follows the block diagram shown in the system description section above This circuit was first created and simulated in OrCAD Once we gained further understanding of other schematic software we chose CadSoft EAGLE Layout Editor to create the final schematic as you see below The MOSFET was chosen carefully as this component was crucial in the ability to handle design specifications pz p
22. idth was approximately 930 ns By modifying the coil parameters it is possible to increase or decrease this measurement May 15 30 Ea One Microsecond Pulse Width Demonstration 1 4 4 2 1 0 8 0 6 Voltage V 0 4 1 50 1 50 04 Time us Figure 20 One Microsecond Pulse Width Demonstration The results using the magneto optic MO material our client wanted to use Integrated Photonics Inc MGL Garnet Latching Faraday Rotator has a switching field gt 500 G at 22 C That tells us that if we can obtain at least a 500 G field at approximatelv room temperature we should see the polarization of the material change thus changing the power measured at the output Using MATLAB scripts created by our client we took a measurement before turning our circuit on another measure after turning our circuit on then combined the data to measure the phase shift to see if it is what we would expect Below in Figure 21 are the results of our initial test showing we experienced virtually no phase shift May 15 30 EN File Edit View Insert Tools Desktop Window Help x OQGASJRkIR sRsVDRAs4 JaDgan Output Power as a Function of Polarization Setting Output Power dB L L 1 1 L L f 0 50 100 150 200 250 300 350 400 Polarization Setting Degree Normalized Output Power a Output Power dB Cy oS Bob gt o o b d 0 1 0 50 100 150 200 250 300 350 400 Pola
23. igi Key 10 490 1312 1 ND o 0 012 0 12 Digi Key 10 399 3677 1 ND EE 0 329 3 29 Digi Kev 4 718 1945 1 ND G 10 88 43 52 Digi Kev 2 ED2675 ND wa GR Seng 1 04 2 08 Digi Key 14 J716 ND Coaxial Connector 4 84 67 76 Digi Key 4 ED10561 ND Wire to Board Connector 0 58 2 32 Digi Key CP 002AHPJCT ND DC Barrel Jack 1 89 15 12 Digi Key N A PCB Fabrication 5 00 30 00 OSHPark Total Cost 274 31 Table 3 Cost of Parts per Board Fabricated Cost Per Quantitv Digi Key Part Number Item Description Total Cost Vendor Unit 1 BZT52C11 FDICT ND 11V Zener Diode 0 21 0 21 Digi Key BZT52C16 FDICT ND 16V Zener Diode Digi Key 0 050 Current 1 MP725 0 050 FCT ND Sense Resistor Digi Key 1 P931FCT ND 9310 Resistor Digi Key 1 P1 43KFCT ND 1 43KQ Resistor Digi Key 1 1276 2740 1 ND 0 01 uF Capacitor Digi Key 2 399 3770 1 ND 100 uF Capacitor Digi Key 1 160 1169 1 ND Green LED Digi Key 2 J716 ND SMA Connector Digi Key 1 568 11554 1 ND MOSFET Digi Key May 15 30 9 APPENDIX E MATLAB CODE 9 1 SINGLE COIL CODE 9 1 1 MAGNETIC FIELD CALCULATION yfiles megsharp b Documents MATLAB gt EE492 B Editor iastate edu cyfiles megsharp Documents MATLAB EE492 BfieldCalculationS hortCoil491 m F CurrentCalculationSingleShortCoil491 m BfieldCalculationSingleShortCoil491 m Los N table Singlem i clear all 2 close all a clc 4 a N inpu
24. nts The resistance affects the functionalitv of the coil as well The resistance lowers the current which also lowers the magnitude of our magnetic field therefore the lower the resistance of the coil the easier it is for the magnetic field to reach 500 gauss Our team did a test by measuring the resistance of a five turn coil with an Agilent 4263 Model B LCR meter and measured resistance at 0 0840 Based on this we expect the final resistance of the coil to have little impact on the final design May 15 30 ll 1 3 6 SINGLE COIL VS HELMHOLTZ COIL CoilA Coil B Coil C Coil D Coil E Frequency Inductance nH Inductance nH Inductance nH Inductance nH Inductance nH 0 5 MHz 73 81 134 01 113 41 294 91 329 01 1 MHz 59 93 119 93 102 83 289 23 307 33 1 5 MHz 68 62 126 22 111 02 282 22 305 42 2 MHz 67 21 124 81 106 91 280 91 296 41 Figure 4 Single and Helmholtz Coils Two different types of coils with dissimilar properties were chosen to use throughout the course of the project a single coil and a Helmholtz coil We tested both to see which yielded the best results and decided on the single coil configuration There is a definite trade off between a single and a Helmholtz coil While a Helmholtz coil configuration generates a more powerful magnetic field with the same coil parameters and same current the single coil has less inductance We have calculated possible coil dimensions including theoretical inductances for both a Helmholtz
25. pply using the DC Power Cable i DC supply must be able to output 15 VDC ii When connecting to the DC supply red is positive black is negative 5 2 2 WAVEFORM GENERATOR AND DC SUPPLY SETUP To properly configure the DC power supply and waveform generator to output what we are looking for 1 Make sure DC supply and waveform generator are OFF 2 Set the DC supply to output 15 V but keep it turned OFF 3 Set the Output Function to Pulse 4 Set the Run Mode to Continuous 5 Set the Output Load to be High Impedance 6 Set the Period 7 Set the Pulse Width 8 Set Low and High Amplitude 9 Set Leading and Trailing Edge 5 2 3 ERRORS AND INDICATION There is one green LED to indicate whether the DC voltage source is on and functioning properly There are also red LEDs to indicate when the voltage is set too high and Zener diodes to redirect a portion of the current once the voltage is increased to a certain point The red LEDs will come on if 1 VDC is increased over 18V while the DC supply is ON 2 VAC is increased over 13V while the waveform generator is ON May 15 30 ES 6 APPENDIX B ADDITIONAL INFORMATION 6 1 RESOURCES Throughout the duration of this project resources were used from lowa State A detailed list is shown below Financial resources for parts tools and supplies necessary to perform the research o Test and measurement equipment software Agilent 34410A Digital Multimeter Tektronix DPO 4032 Digital Phospho
26. r Oscilloscope Tektronix AFG 3021B Single Channel Arbitrary Function Generator Agilent E3630A Power Supply Hewlett Packard 8714ES RF Network Analyzer Agilent 8164A Lightwave Measurement System Agilent 8169A Polarization Controller 1550nm Optical Laser Module Optical Power Sensor Module Lakeshore 410 Gaussmeter Thorlabs DET10C Power Sensor Transimpedance Amplifier MATLAB Software License O O O O O O O O O O O 0 0 6 2 LITERATURE SURVEY This circuit can be used in many applications such as Small scale fiber optic switches and routers Megawatt Q switched laser systems Research in biomagnetism Small solenoid systems Although a single coil field generator design is not a new area few applications require small devices with such high current requirements at high speeds With this said little research has been done to create such a device J W Pritchard M Mina R J Weber 2013 July 7 Magnetic Field Generator Design for Magneto Optic Switching Applications Online Available http ieeexplore ieee org stamp stamp jso arnumber 6558961 May 15 30 ES J W Pritchard M Mina N R V Bouda 2013 Feel the pulse Magnetics Technology International Online Vol 2013 pages 04 07 Available http viewer zmags com publication 17fdeQadH 17fdeQad 6 Tioh Jin Wei Interferometric switches for transparent networks development and integration 2012 Graduate Theses and Dissertations Paper 12487
27. rease the package sizes and make them more uniform Decrease the size of the wire to board connector Efficiently use the space on the board and don t maximize the dimensions if we don t need to use them all May 15 30 ES Quantitv Digi Key Part Number Item Description Cost Per Unit Vendor 10 497 6085 1 ND Diode 0 72 7 22 Digi Key 10 1N4148WTPMSCT ND General Purpose Diode 0 14 1 40 Digi Key 10 MMSZ5233B TPMSCT ND 6V Zener Diode 0 21 2 10 Digi Key 10 BZT52C6V8 FDICT ND 6 8V Zener Diode 0 21 52 10 Digi Kev 10 BZT52C11 FDICT ND 11V Zener Diode 0 21 2 10 Digi Key 10 BZT52C12 FDICT ND 12V Zener Diode 50 22 52 20 Digi Kev 10 BZT52C16 FDICT ND 16V Zener Diode 0 21 2 10 Digi Key 10 a a 17V Zener Diode 0 24 2 40 Digi Key 6 MP725 0 050 FCT ND ba GE SC 9 13 54 78 Digi Key 10 P2 2ECT ND 2 20 Resistor 50 10 1 00 Digi Key 10 P49 9FCT ND 49 90 Resistor 0 10 1 00 Digi Key 10 P422FCT ND 4220 Resistor 50 10 1 00 Digi Key 10 P931FCT ND 9310 Resistor 0 10 1 00 Digi Key 10 P1 3KECT ND 1 30KQ Resistor 50 10 1 00 Digi Key 10 P1 43KFCT ND 1 43KQ Resistor 0 10 1 00 Digi Key 10 160 1169 1 ND Green LED 50 37 53 70 Digi Kev 10 160 1167 1 ND Red LED 0 40 4 00 Digi Key 10 1276 2740 1 ND 0 01 uF Capacitor 0 13 1 30 Digi Key 10 399 3677 1 ND 0 1 uF Capacitor 0 38 3 80 Digi Key 10 399 3770 1 ND 100 uF Capacitor 1 49 14 90 D
28. rization Setting Degree Figure 21 Output Power Phase Shift Measurement At this point we knew either our circuit was not performing as intended or the MO material was not reacting as expected This led us to develop a new test 3 6 3 FINAL TEST SETUP A new method was developed to demonstrate the functionality of our circuit that involved us making theoretical calculations on the field strength we could obtain by reducing our input parameters then performing a test on a magneto optic MO material that could be affected by a smaller magnetic field strength We would then use that data to determine if 500 G was attainable by scaling the parameters to their maximum values The new MO material was the Integrated Photonics Inc FLM Garnet Low Moment Faraday Rotator This material operates differently than the MGL latching material does in the respect that it will reach a saturation point when the magnetic field strength exposed to it is 2 225 G Using our magnetic field calculations and modifying our parameters we were able to determine what we can set the parameters to in order to see the MO material saturate Our measurements would be taken on the voltage across the current sense resistor CSR as well as the change in power of the optical signal while we pulse our magnetic field To do this we needed to increase the pulse width of May 15 30 EI our signal Our expectation was that once the MO material was saturated the ch
29. t After testing we figured out we had installed the barrel jack incorrectly and the pins were mixed up To solve our problem on current versions of our board we simply jumped pin A and pin B together as we did not need the functionality of the non grounding pin This issue is fixed on future layouts Figure 25 DC Barrel Jack Schematic 4 CONCLUSION The goal of this project was to create a magnetic field generating circuit that could create a one microsecond pulse with a 500 gauss magnetic field Our team s collection of circuit knowledge EMF understanding design process mindfulness among other skills learned along the way is shown in this descriptive document This project was successful in providing the necessary requirements for the client as outlined in the 3 5 Data Analysis section of this document We have completed this project with the utmost confidence the magnetic pulse generating device will be used in future applications May 15 30 EN 5 APPENDIX A USER MANUAL 5 1 OVERVIEW 5 1 1 COMPONENT LIST Table 1 Component List CL 100 uF Capacitor C2 100 uF Capacitor c3 0 1 uF Capacitor C4 0 01 uF Capacitor D1 16V Zener Diode D2 6V Zener Diode D3 6 8V Zener Diode D4 17V Zener Diode D5 100V Diode LED1 Green LED LED2 RED LED LED3 RED LED R1 20 Resistor R2 500 Resistor R3 1 43kQ Resistor R4 1 30kQ Resistor R5 4220 Resistor May 15 30 EN 5 1 2 PCB
30. t How many turns are in the coil N gt of turns in the coil dimensionless Ge 1 input What is the length of the coil in mm 1 gt length meters a R input What is the radius of the coil in mm R gt radius of the coil meters LE I input What is the current of the coil I gt current amperes 9 I5 BT 4 pi 10 7 N I sqrt 1 0 001 2 4 R 0 001 2 Equation from MFG Design paper for single coil EE B g B_T 10 4 Converting magnetic field units from Tesla to gauss 12 disp The magnetic field generated with those parameters for a single short coil is num2str B_g Gauss 14 9 1 2 CURRENT CALCULATION yfiles megsharp b Documents b MATLAB b EE492 Editor iastate edu cyfiles megsharp Documents MATLAB EE492 CurrentCalculationSingleShortCoil491 m ji J CurrentCalculationSingleShortCoil491 m gt BfieldCalculationSingleShortCoil491m Luc N table Singlem 1 clear all 2 close all s clc 4 Zi N input How many turns are in the coil N gt f of turns in the coil dimensionless e 1 input What is the length of the coil in mm 1 gt length meters Ta R input What is the radius of the coil in mm R gt radius of the coil meters re B_g input What is the magnetic field in Gauss of the coil Bg gt magnetic field produced by the coil Gauss 9 20 BT B g 10 4 Converting magnetic field uni
31. ts from Gauss to Tesla 11 I B_T sqrt 1 001 2 4 R 001 2 4 pi 10 7 N Equation from MFG Design paper for single coil 12 33 disp The current required to generate that field for a single short coil is num2str I Amps 14 9 1 3 PARAMETERS OF 500 GAUSS CURRENT INDUCTANCE VS NUMBER OF TURNS May 15 30 te edu cyfiles b megsharp b Documents MATLAB b EE492 Editor iastate edu cyfiles megsharp Documents MATLAB EE492 I_vs_N_table_ Single m woody nn Go bt SN P NM RRR NNR eRe ore roe DD A o m WNP OW DD A o WO Ga h O CurrentCalculationSingleShortCoil49Lm BfieldCalculationSingleShortCoild9Lm X Los N table Singlem 1 input What length do you want to work with num rad input How many radii do you want to enter try x num_rad 0 sfor k 1 1 num rad r input What radius do you want to graph in mm tend N 1 1 20 E zeros 1 20 L zeros 1 20 sfor j l num rad for i 1 1 20 I i 0 05 sqrt 1 0 001 2 4 r 0 001 2 4 pi 10 7 N i L i 1000 4 pi 10 7 N i 2 pi r 2 sqrt 1 0 001 2444 rx0 001 72 end figure j grid on hold on hAx hi h2 plotyy N I N L str sprintf Turns vs Current Single Short Coil R tfmm r title Single Coil 500 Gauss Parameters xlabel of Turns txlim 0 50 ylabel hAx 1 Current Amps tylim 0 100 ylabel hAx 2 Inductance
32. tures roeier n n a a EE EARE EEE E iin RES 21 3 6 Data Zuele tee ee EN Mav 15 30 ES 3 6 1 initial Fest d sine dis ninie rE EEE EEEE EE EESE 21 3 6 2 Initial Data tin ka i i fe Ann ei 22 3 6 3 Final Fest idi gie a 24 364 Final Data doy esp saps teense eng cea E AE E e E ainsi T 25 Re Kg E 25 3 7 Implementation Issues Challenges sssessenenzanzznzzznnnaninzznnznznnzinnaanenazznnzntnz anna nznnanzna 26 3 7 1 Soldering Small COMPONENTS i ib ise Ae gereent eege 26 3 7 2 Traces Lifting Oi a ccctuacsstagiieseupautctladgatictedadus ina en a idi kosta aae iaa 26 3 7 3 High Current Risk of Burning Components 27 37 4 ee E 27 3 7 5 D Bal fel Jack Ee 27 a E ele 28 b Appendix A User Manual getest gegen eege be eg 29 E E ae EE 29 5 1 1 eine ele lt 29 51 27 BERCHEM 30 Did SEW ira a a a a 31 5 2 1 Eeer 31 5 2 2 Waveform Generator and DC Supply Setup 32 5 2 3 Errors and Indi atioti siket sinifika kein e wa ai 32 6 Appendix B Additional Information 33 ENN le 33 6 2 Literature SUMCY e dE EEN dE SEN 33 6 3 E elen E 34 GW e EE 34 7 Appendix C First Desi aerial 35 8 Appendix D Bill of E E 36 9 Appendix E MATLAB Code ssssnsismiissciriesericssirreionianaedsnere sssri sikio jaU 38 9 1 Single e Re TE 38 Mav 15 30 ES 9 1 1 Magnetic Field Ealewlatlep Sessoeuiegegedertek Geseent egeEEAe egene 38 9 1 2 QirrentCalcenatioti E 38 9 1 3 Parameters of 500 gauss Current inductance vs number Of ture 38 92 Helmhol
33. tz cod e a a 39 9 2 1 Magnetic Field Calculations x sacctocsctsansecvsrcgnentd nip aceencdenset lnevaavscaciansloncondveantenaabenneedehs 39 9 2 2 CHAE CANCUN EE 40 May 15 30 ES l SYSTEM LEVEL DESIGN 1 1 SUMMARY OF DEVICE As communication technology utilizing new magneto optic materials advances there becomes and increased need for small scale high powered magnetic field generators Although a single coil field generator design is not a new area few applications require small devices with such high current requirements at high speeds With this said little research has been done to create such a device This senior design project presents a magnetic field generation circuit to be used in a research grade fiber optic switch with magneto optic material This document covers the overall design strategy functionality applications and results of this magnetic pulse generator project Figure 1 Electromagnetic Pulse Generator Prototype 1 2 PROJECT REQUIREMENTS 1 2 1 FUNCTIONAL REQUIREMENTS The device shall generate a magnetic pulse of 500 gauss or greater The duration of the pulse shall be one microsecond 1 2 2 NON FUNCTIONAL REQUIREMENTS May 15 30 ES The device shall be in an enclosure readv The device shall have a DC source voltage of 15 volts or less The device shall not exceed the dimensions 3 5 L x 2 0 W 1 3 FUNCTIONAL DECOMPOSITION 1 3 1 POWER SUPPLY Our design calls for th
34. z EA yy Figure 9 Circuit Schematic May 15 30 Fe 3 3 2 SIMULATION RESULTS After a basic circuit schematic was decided upon the team moved to simulate the circuit using OrCAD software Our first step was to decide on a MOSFET The MOSFET that chosen based on fast switching speed to accommodate the one microsecond pulse width and be able to withstand a high current to generate the required magnetic field Below figure 10 shows two MOSFETS that we were considering using in our circuit The MOSFET spice files were obtained directly from the manufacturer s website V120 V120 V2 5 V2 5 TD 0 TD 0 TR 0 TR 0 TF 0 TF 0 PW tus PW tus PER ims PER ims Figure 10 MOSFET Considerations Once the simulation was ran we could see the individual results for the MOSFETs as how they react to the one microsecond pulse in conjunction with the coil As you can see in the Figure 11 below the left MOSFET performed better than the right one We can see that in the faster rise time followed by the shorter fall time May 15 30 e 15A PSMNOR9 30YLD IRL3714Z_S_L old BA 8s 8 5ms 1 6ns 1 Ems 2 0ms 2 5ms I L2 I L1 Time Figure 11 MOSFET Performance After the MOSFET was chosen the simulation team worked with the coil team to determine the most appropriate coil to use Once simulations began it quickly became apparent that a smaller inductance in the coil would yield a smaller rise time This is shown
Download Pdf Manuals
Related Search