reproducibilitylimitswereobtainedfornon-standardtestmethodsaftercorrectingtheinstrumentbetweendifferenttimeintervals.Basedonthestimulatedrepeatabilitylimitandwithin-laboratoryreproducibilitylimit,thelong-termstabilityofthecorrespondinganalyticalinstrumentmethodcanbeevaluated.Thispaperusesthenon-standardmethodfordetermining18impurityelementsinpurenickelbyglowdischargemassspectrometry(GD-MS)asanexample.Asetofcomparativeexperimentsaredesignedtoobtainthesimulationreproducibilitylimitandsimulationreproducibilitylimitdata.Thelong-termstabilitytimefor18impurityelements,suchasB,Mg,AlandSiinpurenickelsampleswasevaluatedbyχ2statisticsafterbeingtestedbyGD-MS.Theresultsshowedthatthelong-termstabilitywasdifferentfordifferentelementswhendeterminedunderthesamecondition.Theresultsmetthestatisticalrequirementsformostelementswithin3hours.Amongthese,thelong-termstabilityintervalcanbeupto6hoursoreven12hoursforelementsP,As,V,SbandPb.Thereliableresultscanbeobtainedwithinthattimeintervalwithoutinstrumentco...