Gas Detection Strategies,Overview of Selection Criteria for Gas Detection Systems,How do I choose?,Which Gas Detection Technology is best for my application? What criteria is important to me? How do I understand the specifications that are given to me?,Which Technology is Best for my Application?,Understand that there are different technologies for most gases Understand that some technologies do not work for certain gases Examples: FTIR cannot detect diatomic molecules (Cl2, Br2) Example: Paper Tape (Chemcassettes) cannot detect Hydrogen,Which Technology is Best for my Application?,Develop a selection criteria and use it to evaluate the different technologies Make sure you are comparing apples with apples many suppliers call data by the same name but calculate it differently,Developing Selection Criteria,Questions to Consider Before We Begin Review Fab Monitoring Locations - Tool Exhaust - Fab Ambient Air - VMB - Gas Cabinet - Subfab Ambient - Pump Housings - Process Protection Applications Do You Want the Same Criteria for Each Location?,Developing Selection Criteria,What is Important for the Application (Usually Same List / Sometimes Different Priorities) Specificity Speed of Response Lower Detectable Limits Reliability Accuracy Cost of Ownership Initial Cost of Capital Investment,Developing Selection Criteria,Establish a Spreadsheet for each Monitoring Application (Ambient, Exhaust, Gas Cabinet, etc.) List your Criteria Rank your Criteria Sort your list by Rank Evaluate each technology by the criteria Choose the technology that best meets your needs,Developing Selection Criteria,Developing Selection Criteria,Developing Selection Critera,You May Ask Yourself. Specificity? Speed of Response? Reliability? Accuracy? Lower Detectible Limits? What Does Do These Terms Really Mean to Gas Detection?,Specificity,No cross sensitivities to other substances commonly used in a semiconductor fab Review manufacturer data this information should be published and readily available Be sure to include non toxics in your review such as cleaning solvents! They may have undesirable effects on your gas detection system.,Speed of Response,Speed of response is the most difficult data to compare between technologies Several factors affect speed data Data often stated differently from technology to technology, makes comparison difficult Hidden technical aspects unique to each technology,Speed of Response,Statements of Response 30 seconds at TLV Response time usually varies over the range of the technology, slower at lower concentrations, faster at higher concentrations T90 in 30 seconds Sensors responds to 90% of the concentration in 30 seconds, hits a response peak, levels out and very slowly responds to the last 10%. If the leak concentration borders around the TLV, you may only detect 90% of TLV, and not actuate an alarm,Speed of Response,Statements of Response Other Factors to Consider Some technologies respond to a leak at the end of a fixed period Some have to verify the leak if other cross interferences are present, adding to the time to alarm Many technologies do not have the same response time over the lifespan they are quick when first installed but become slower, more sluggish the longer they are installed in the field. When recalibrated at specific intervals, concentration response is measured but many test companies do not log the time to response, showing that the sensor is responding much slower than expected.,Speed of Response,Statements of Response Other Factors to consider (continued) Some technologies are not continuous, so the time it takes to scan between points needs to be added in Beware of suppliers who tell you their response is “instantaneous” that is NOT data! Ask them to define instantaneous with data. You will find it means 10, 20, 30 seconds.,Lower Detectable Limit,LDL is important to semiconductor industry due to high air flow rats High air flows can dilute gases to a level undetectable by some technologies No monitor starts at “zero” every has a level between zero and LDL which is electronic noise where they cannot reliably detect. Question 0-100 as a monitoring range.,Reliability,High air flows can dry out /cause false detections on some technologies Cross-sensitivities Drift (caused by temperature, humidity, air flow) RF interference can cause false positive detections,Reliability,Look for extensive self-diagnostics that can compensate for known problems Can the detector compensate for drift and translate it as a maintenance issue not a false gas alarm? Can the detector identify problems that could cause it to miss a gas detection (paper tape out of tape, electrochemical dried or failed sensors, FTIR spectrums overlapping the target gas detection wave length area),Accuracy,Gas detection should have a stated accuracy specification Example: +/- 10% the direct reading Example: +/- 10% at TLV These statements do NOT mean the same thing!,Accuracy,Example: minimum of 2% f