YIELDANALYSIS
methodology involves automatically generating sample data based on between five and twenty known good runs, the data sample created is referred to as an envelope.
This envelope is then used in the same way as upper and lower limits as a comparator for subsequent process runs, to either provide an alarm or hold the processing of subsequent batches if out of an tolerance event is observed.
Critical conclusions For PV manufacturers the diffusion processes are critical to the performance of the device and in particular the doping process. For a conventional POCl3 or BBr3 doping process, temperature and time or the overall thermal budget, in conjunction with repeatable gas flow are the critical variables in determining the emitter junction depth and uniformity.
Having a data collection service to record all process information including temperature, gas flow, element power, etc. and then using SPC control charts to monitor and identify out of tolerance variables is a very powerful tool in ensuring cell performance and avoiding hardware failure and consequent downtime.
Typical software packages may also provide pareto anlaysis of alarms and system events, helping engineers to understand where there may be any recurring issues and pinpointing the causes.
In conclusion, SPC software programs will become more and more important to PV manufacturing as the critical production processes become better understood. The quality of the production equipment determines the quality of the end product and SPC is a vital resource, as semiconductor manufacturers can attest, in maintaining the quality of the equipment.
© 2011 Angel Business Communications. Permission required.
Some typical SPC calculations performed on the raw data Min
Max Start End
Average (Mean)
Minimum value seen during collection Minimum value seen during collection First value seen at beginning of collection Last value seen during collection
The simple average of the data seen during collection
Standard Deviation (Sigma) The simple standard deviation of the data * Standard Deviation % * Cp
* Cpk
* Range Dt
Dt e2/kT
Signature * Slope
* Duration Standard Deviation / Mean as a percentage Calculated as (User High Spec – User Low Spec) / *6 * StdDev)
Calculated as Cp – ABS( ( (UserHig + UserLow) / 2 /Mean ) / (3 * StdDev) ) Calculates and uses (Max – Min)
Simple Integral of (reading * time sample) calculated during the data collection for the Period, where the time units are minutes.
Calculated during collection as 1000 * Log of ( Sum( period * Exponential of ( value / 1000) ) ) where the time unit is minutes.
(test is based on the overlay data envelope) Calculated as the (End – Start) / Duration
Time from start of collection to time the last sample collected
* These items are calculated on display of the Control Chart from Data Period collected data. All other fields are generated as part of the Data Period collection.
45
www.solar-pv-management.com Issue VII 2011
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56