Testing and Validation
Contents
1. the driver is manually calculated through Ohm s Law using the measured voltage across the resistance of the wire by the oscilloscope This current value is then compared to the displayed current on the LCD of the controller These two numbers should match within half of an amp of each other When the controller is functioning correctly with no errors the setup is left to run for the rest of the day and the following next couple days If no errors occur during this testing period it is concluded the controller is stable enough to connect to the actual laser diode After it is installed into the system thorough testing of the controller on the system is performed by the researchers building and tuning the laser Results Currently there have been two completed controllers that are functioning on two different laser diodes The only issue found coming from testing of the completed controller is sporadic tripping of the interlocks in the lab After further diagnosis of the problem the noise coming from the lab was randomly triggering the interlock by creating small spikes in the voltage in the traces This problem was fixed by adding a 56 pF capacitor to each of the inputs on the interlocks No issues have occurred since2. Laser Diode Control System Caorado BUY University An NSF Engineering Research Center Team Members Ryan Evans Senior Design Electrical Engineering BS Brandon Carr Electrical Engineering BS Supervising Professor Dr Jorge Rocca Summary In the Extreme Ultraviolet Engineering Research Center EUV ERC diode pumped lasers are being used to create a soft x ray laser that will be used in many future applications Diode pumped lasers require a controller to manipulate the voltage and current the diode receives Currently there are controllers that provide a regulated voltage and current to supply the diode with enough power to create a laser The goal is to implement a new updated diode controller to reduce the effects of high frequency noise and increase the amount of protection on each of the controller circuit boards New features software and box layout will be implemented along with possible redesigns of board level circuitry The final goal includes installing five stable noise resistant controllers to control ten different laser diodes Importance of Project The initial goal set by Dr Rocca was to create a diode pump soft x ray laser with a 1 kHz repetition rate At this high of a repetition rate fast switching becomes an obstacle to the current controller due to an increase in electromagnetic noise from not only the controller itself but also all of the components in the laser system The controller will be able to
3. an interlock does not trip on the display and the rest of the LCD is functioning correctly there must be a break in the trace an issue with the processor or an issue within the code Testing the outputs of the controller to the Quasi Continuous Wave driver QCW or Laser Diode Driver High Power Pulser LDD HPP is one of the most critical steps of the entire testing process The DACs are the devices on the board that output a voltage that tells the QCW or LDD HPP what current to output to the laser diode The enable and ready signals are applied in the code and the rotary encoder is adjusted up and down while a multimeter is placed on the output of the controller The voltage should vary linearly with the increase in current from 0 10V If this does not happen a fault in the code or a bad connection could be the issue A multimeter is used to accurately measure the output voltage from the DACs Once this test is completed it is concluded that the board is ready to be installed in the controller enclosure To test the rotary encoder a multimeter and the LCD were needed The LCD was used to make sure that when the minimum or maximum value of the current was reached the encoder did not go above or below the value or start over This could also be seen through the multimeter At its minimum the output value of the current should be OA and at its maximum value the current should be set to 200A This test is similar to the testing of the DACs previo
4. andon Carr 7 31 Completed 7 31 Assemble box with all wiring Performed by Ryan Evans 8 5 Completed 8 3 Finish testing of controller on dummy load Performed by Ryan Evans and Brandon Carr 8 1 10 Completed 8 6 Test controller on system and install on laser diode Performed by Ryan Evans and Brandon Carr 8 15 Completed 8 8 Completely build test and install LDD HPP laser diode controller Performed by Ryan Evans and Brandon Carr 9 20 Completed 9 23 Develop software for rotary encoder potentiometer Performed by Ryan Evans and Brandon Carr 10 10 Completed 10 23 est rotary encoder on newest OCW boards Performed by Ryan Evans 10 24 Completed 10 28 Completely build test and install QCW laser diode controller Performed by Ryan Evans and Brandon Carr 11 21 Completely build test and install LDD HPP laser diode controller Performed by Ryan Evans and Brandon Carr 12 19 Spring 2015 Completely build test and install LDD HPP laser diode controller Performed by Ryan Evans and Brandon Carr 2 20 Remove 1 QCW controller and updated modifications Performed by Ryan Evans 3 6 Remove 1 LDD HPP controller and update with new modifications Performed by Ryan Evans 3 20 Document a user s manual and technical detail Performed by Ryan Evans 4 30 Device Testing and Validation Circuit Board Design Requirements Static shocks and shorting if electr
5. handle the increased repetition rate while also being almost immune to the noisy environment Problem Statement Laser diodes require very precise voltages and currents to create a laser An unexpected shock to the boards or slight variation in the output current can result in damage of very sensitive and expensive equipment The greatest concern is keeping any of these scenarios from happening Objectives The objective is to design a controller that can precisely control output and simmer currents while also being able to monitor the status of different components within the system Controller stability is key in the design If any hardware or software fails because of interference from noise inside or outside the box valuable equipment could be damaged and destroy key components within the system There are two systems that these controllers will be manipulating Quasi Continuous Wave QCW and Laser Diode Driver High Powered Pulser LDD HPP systems Figure 1 shows how the entire system will come together Pulsed DC Voltage Laser Diode 120 VAC 15 Pin Connector Chiller Controller Controller Inputs Figure 1 QCW Laser Diode System The controller that will be developed and assembled will power the QCW while also controlling the output current the laser diode sees Chillers are used to cool parts of the laser system Although the QCW controller will not be powering the chiller
6. ical components on the board will not damage other components in the laser system There will be no loose unsoldered pins or shorts created by extra solder on the pins 5V and 15V will be supplied to all of the correct pins on the board Testing of the PCBs used in the diode controllers is a critical step in producing a product that will be used for years to come The first test done to the PCB is to make sure all of the pins after surface mounting are making good connections to the board A dental pick will be used on the buffers diodes digital to analog converters DACs and the processor to make sure a solid solder joint was made Testing the impedance between the power plane and the ground plane is an easy way to determine if there are any shorts between them A short between these two planes will lead to failure of the entire board Once the testing of hardware components on the PCB is done the next step is to power the board upload the software and begin to measure the outputs from the board to see if the inputs from the software match the outputs One 5V power supply and one 15V power supply are mounted within the controller box to power the board Vias were designed into the board layout to allow easy access to critical test points within the circuit Each via s voltage is measured to ensure the correct voltages are being supplied to the correct rails within the board If shorting between the planes occur finding the short in the so
7. in the lab where it will be used Instead of connecting the controller to the laser diode before final testing has been completed an array of smaller diodes is used to imitate the voltage drop associated with the actual laser diode The array of diodes is soldered in series on a small through hole testing board A QCW or LDD HPP driver must be used at this point in testing because it must be shown that output current from wither of these drivers can be controlled by the controller accurately A multimeter is placed across the diode array to measure the output voltage of the driver and an oscilloscope is connected across a solid wire with about a 0 25 Ohm resistance to plot the output of the current pulse coming from the driver The pulse must look clean without any bouncing after its rise and fall while also not having any variation in current from external noise in the lab One of the reasons this test is done in the lab is because the amount of noise the controller will be exposed to Electromagnetic interference from the high powered switching and pulsing can create problems for controller electronics in the lab which is one reason why the diode controllers needed to be revised The controller is not able to pulse the amount of current necessary to power the laser diode because the diodes used would not be able to handle the amount of power A lower frequency pulse at a lower current must be used so the diodes do not burn Experimental output current of
8. it will be connected to it via an interlock If there are any errors in the flow or level of the chiller the interlock will trip and cause the system to shut down Signals are sent to and from the controller and QCW through a 15 pin DSUB connector Other interlocks for different parts of the laser system will be connected to the controller for further protection of the system Figure 2 illustrates the basic design and implementation of the controller in the LDD HPP laser system 3 Phase Pulsed DC Voltage Laser Diode 15 Pin Connectors Optoisolator Chiller Controller Controller Inputs Figure 2 LDD HPP Laser Diode System The LDD HPP controller uses the same hardware as the QCW system but the difference lies in the code Because this system requires two components to create the pulsed DC voltage different code must be used to control each entity The LDD and HPP must be separate in this system because the LDD is not able to pulse the amount of power it is outputting fast enough Instead of the controller powering the diode driver and the pulser it will power an external relay that will enable the LDD This system uses an optoisolator between the controller and the HPP to create two separate ground planes The optoisolator must be used to protect both incoming and outgoing signals from the electromagnetic noise created by the fast switching of the HPP Final Design Info
9. lder joints must be done before any other testing occurs The voltage is measured by a multimeter to determine if there are any unwanted drops in voltage between the planes or traces If there is an issue the trace will be followed and all of the voltages across all components on the trace will be checked to make sure any component failures are fixed before further testing occurs Software Hardware Design Requirements All interlocks will completely disable the laser system if any of them are triggered The voltage outputs from the digital to analog converters will output 0 10V to the QCW or LDD HPP The rotary encoder will increment the driver current and set current in divisions of 0 05A per state change at its maximum it will not go higher than 200A and at its minimum it will not go lower than 0A The software is easily uploaded to our processor through our JTAG interface With the software outputs from the controller will be measured with a multimeter and compared to our software inputs One of the first tests completed is the functionality of our interlocks Interlocks will disable the entire system if any of them trip To trip an interlock all that needs to be done is to short the connection If the number of the interlock disappears on the LCD when the connection is shorted the interlock is working correctly When the seven different interlocks are tripped and functionally working the testing of the interlocks is complete If
10. rmation LCD s display values of output current simmer current and flow Tripping one of the four interlocks will disable the voltage going to the QCW or LDD The controller will not be affected by any interior or exterior electromagnetic noise Output currents will be accurate with output DAC voltages Controller will be able to power both circuit boards the QCW a chiller and or an external relay All controller boxes will be visually identical LCD s will be able to display interlocks when one is tripped Design Constraints User friendly to almost anyone At least four interlocks per controller Controllers should be stackable or mountable to a rack Output currents should be 200A on the QCW and 180A on the LDD HPP Interlocks malfunctioning Inaccurate current readings Damage to laser diodes Noise from system components interfering with board signals Spikes in voltage on the PCB boards Risk Mitigation Techniques Applying filters to interlock inputs and all other inputs to decrease noise Adding digital and analog buffers Inserting optoisolators Pull up and pull down resistors on almost all power rails Decoupling capacitors Testing more testing and even more testing of all situations Timeline Fall 2014 Solder two OCW controller boards Performed by Ryan Evans 7 28 Completed 7 29 Test both boards outside of box Performed by Ryan Evans and Br
11. usly but this time the voltage values between 0 10V had to correspond linearly to the values displayed on the LCD Assembled Controller There are a couple different stages in testing once the controller is fully assembled To check the wiring and connection of all of the terminals a multimeter is used to measure the different voltages at different points within the controller to make sure all voltages are correct The same output test that was completed previously on the PCB is completed with the assembled controller Interlocks are completely checked for loose connections or faulty cables the LCD is checked for any output errors and all of the same outputs from the controller are rechecked to make sure the wiring did not affect any of the outputs or create any unwanted shorts in the box Load Based Design Requirements The recreated current pulses will have a rise time and fall time less than 6 8 ns and external and internal electromagnetic noise will not interfere with the pulse s profile and stability Displayed current values on the LCD correspond directly to actual output current values through the 0 25 Ohm resistance Internal and external electromagnetic interference EMI will not interfere with that functionality of the controller when implemented in the high power fast switching environment within the system Once it is shown that everything is working on the controller a real time test of the controller is completed
Download Pdf Manuals
Related Search