Jump to content

Specific Process Knowledge/Characterization/Profiler: Difference between revisions

← Older editNewer edit →
VisualWikitext
Reet (talk | contribs)
DCH-Employees-701, NLAB-Employees-701, NLAB-LabmanagerAllUsers, NLAB-Nlab347-labadviser-users-35231
1,745 edits
Reet (talk | contribs)
DCH-Employees-701, NLAB-Employees-701, NLAB-LabmanagerAllUsers, NLAB-Nlab347-labadviser-users-35231
1,745 edits
Line 93: Line 93:
[[File:Error probability distributions sep 2020 update big.png|upright=2|alt=Four different probability distributions that contribute to the total error on the Dektak measurement for the 6.5 micron range. By far the widest distribution is the one from the error on the calibration standard, which is a Gaussian. The others are the non-Gaussian spread of the average measurement of the calibration standard height, which cuts off at the QC limits, the resolution, which is a very narrow uniform distribution, and the spread of measurement values for a given step being measured, which is a Gaussian whose width depends on the step in question.|right|thumb|The probability distributions of the main sources of error that are convoluted to create the total error on a Dektak measurement.]]
[[File:Error probability distributions sep 2020 update big.png|upright=2|alt=Four different probability distributions that contribute to the total error on the Dektak measurement for the 6.5 micron range. By far the widest distribution is the one from the error on the calibration standard, which is a Gaussian. The others are the non-Gaussian spread of the average measurement of the calibration standard height, which cuts off at the QC limits, the resolution, which is a very narrow uniform distribution, and the spread of measurement values for a given step being measured, which is a Gaussian whose width depends on the step in question.|right|thumb|The probability distributions of the main sources of error that are convoluted to create the total error on a Dektak measurement.]]


Apart from the error due to the calibration standard's uncertainty, there will be random noise in any measurement, which we have found for many repeated measurements of the standard step height (a rigid, well defined vertical step) is on the order of ± 5 nm. There is also a tiny contribution to the error from the instrument's resolution and a small but significant contribution from the spread of values that the Dektak actually measures compared to the theoretical height of the standard step height.  
Apart from the error due to the calibration standard's uncertainty, there will be random noise in any measurement, which we have found for many repeated measurements of the standard step height (a rigid, well defined vertical step) is on the order of ± 5 nm for the 917 nm standard and 0.05 µm for the 24.925 µm sample. There is also a tiny contribution to the error from the instrument's resolution and a contribution from the spread of values that the Dektak actually measures compared to the theoretical height of the standard step height.  


To estimate the overall accuracy of the Dektak's measurements you can convolute the various sources of error. This is shown graphically on the right. You can see an uncertainty budget for the Dektak measurements here (made by Rebecca Ettlinger): [[Media:uncertainty budget Dektak rev.xlsx]]. It is based on the assumption that all the error sources are independent and can therefore be added by the sum of squares method.
To estimate the overall accuracy of the Dektak's measurements you can convolute the various sources of error. The error sources to be convoluted are shown graphically on the right. You can see an uncertainty budget for the Dektak measurements here (made by Rebecca Ettlinger): [[Media:uncertainty budget Dektak rev.xlsx]]. It is based on the assumption that all the error sources are independent and can therefore be added by the sum of squares method.  
The resulting error calculation for a 1 micron very well defined standard step is about 2 % (as the uncertainty on the calibration standard dominates), while for a very well defined step of 25 microns the cumulative error is about 0.7-1 %. These are the uncertainties listed in the table. However, in real devices the random error will often be much larger than for our standard samples and so the real confidence interval will be larger.  


To improve the accuracy of your particular measurement, you should repeat the measurement several times and estimate the standard deviation. If the scatter is quite small you can try to include the calibration error as a percentage of the step height in your estimate of the total error. If the scatter of your measurements is large that will probably be the dominant source of error in your measurement.  
The resulting error calculation for a 1 µm very well defined standard step is about 2 % (as the uncertainty on the calibration standard dominates), while for a very well defined step of 25 µm the cumulative error is about 0.7-1 %. These are the uncertainties listed in the table. However, in real devices the random error will often be much larger than for our standard samples and so the real confidence interval will be larger.
 
To improve the accuracy of your particular measurement, you should repeat the measurement several times and estimate the standard deviation. If the scatter is quite small you can try to include the calibration error as a percentage of the step height in your estimate of the total error using the sum of squares method as done in the uncertainty budget above. If the scatter of your measurements is large that will probably be the dominant source of error in your measurement.  


<br clear="all" />
<br clear="all" />
Retrieved from "https://labadviser.nanolab.dtu.dk//index.php?title=Specific_Process_Knowledge/Characterization/Profiler"