SETTING UP TO PAIR MATCH
There has to be relevant data in nk1_crv.dbf for NK1 Xtra to pair match a particular model. You generate the relevant data by setting the following sequence parameters for the models you wish to pair match. Set Serial Number on to Pass. Set Disc Log on to Pass.
The response stored in the file nk1_crv.dbf is the swept range and this is used for pair matching. Consequently it is IMPORTANT that the swept range should closely match the check range or otherwise unlikely matches will occur. Similarly if the response falls below the bottom of the graticule then noise will be matched against noise, again with unlikely results.
EXECUTING A PAIR MATCH
At the (T)ypical, (P)air Matching, (M)aintenance or (Q)uit prompt select (P)air Matching. You will be presented with a list of models for which relevant data exists in the file nk1_crv.dbf. When a model is selected NK1 Xtra examines the data base reporting the serial numbers, test dates and times of the first and last units found.
The total unit number found is also reported. If the total is less than the value set for Pair Match start population in the system MAINTENANCE menu then you receive an error message and the matching process ends. If the total is greater or equal then the user is given a (C)ontinue or (Q)uit prompt. If (Q)uit is selected then the pair match process ends.
If (C)ontinue is selected then the pair match process starts. Activity is indicated by various text displays during the process. At the end a list shows which units have and have not been matched. When RETURN is pressed a (D)isplay, (P)rint List or (C)ontinue prompt appears. Selecting (D)isplay shows the list again. Selecting (P)rint List sends the list to the current printer.
When (C)ontinue is selected the further prompt Delete source data (Y/N) appears. If you respond Y then you are further prompted Delete the matched drivers ONLY (Y/N). If you respond Y then all the matched units from the file nk_crv.dbf. If you respond N then the whole batch is deleted.
Deleting the source data is necessary to prevent the batch being included when further matching is done. On the other hand you may not delete the data because the engineer wishes to change the relevant system parameters and then re-match the same batch. When first starting up some iteration between adjusting the parameters and matching may be needed to get the required yield.
HOW NK1 Xtra PAIR MATCHES
NK1 Xtra uses simple averaging to pair match. For each unit the y axis values at each frequency are summed and then divided by the number of frequency points. This figure for the first unit is compared with every other unit and the unit with least difference identified. If both the average and worst case differences between the two units are less than the relevant limits then the units are considered to be matched. Both are removed from the matching population. If the pair are not a match then the first unit only is removed from the matching population.
The process continues until the number of units specified by the engineer remain or no more matches can be found. These are left to be included in the next batch to be matched.