中文| EN
0086-15106177808
News

Long-term stability tests on spectrometer

     Long-term stability, mainly caused by instability of CCD when exposed to radiation, is determined by measuring a typical sample 300 times within 8 h. The 30 mean values determined from the 300 values, based on the present state of the art which has not noticeably improved during the past 15 years , are within 2% ie:


`Xmax-Xmin

=                  *100%≤ 2%

X




where

`X= mean values for every 10 measurements

= =mean value for all measurements

X

       Long-term stability tests on spectrometer with SDAR for metal analyses show that the instability of the intensity ratios cannot be assigned either to an element or a reference line, which points to individual behaviour of CCD. Long-term stability can be improved by selecting CCD with similar long-term behaviour.


       Instability amounting to 2% over 8 h cannot be justified for analysis of high-alloy qualities since 2% in the case of 25% is equivalent to is equivalent to D0.5%.Instability does not follow simple functions with time, so that it is not possible to use corrections based on slope and time. Because of the random changes in measurement sensitivity, monitoring of AE spectrometers can only be kept under control by means of check samples. If the check sample is used for type recalibration , accuracy as known from XRF spectrometers can be obtained under repeat  conditions, as shown in Table 8.26c. In this case, the type recalibration sample SUS-CFe3 was measured at different times over 4 days and directly after this analysis samples A and B were analysed. Repeatability is within 0.1% for Cr(-20%) and Ni (-10%). The relative error for Cr is at least 2 times better than for Ni because the relative standard deviation is at 2 times better.

      For the alloy elements accuracy better than ±1% rel2scan be expected under repeat conditions.

      Manual type recalibration is only possible for a limited number of samples eg. This number is also determined by the variety of alloys. The estimated time required for type recalibration is:


Grinding:  1min

4 sparking : 2min

S      : 3min type recalibration

Example for type-recalibration of CrNi-steels Type-recalibration samples: SUS CFe3

Date

Time

Sample A

Sample B

% Cr

%Ni

%Cr

%Ni

Day 1

   18.34

9.71

18.10

9.16

Day 2

18.24

9.66

18.03

9.12

18.28

9.68

18.11

9.16

18.31

9.64

18.05

9.16

Day 3

18.25

9.68

18.04

9.17

18.33

9.63

18.10

9.15

18.26

9.71

18.05

9.18

Day 4

18.28

9.66

18.04

9.14

18.34

9.62

18.14

9.13

18.29

9.70

18.08

9.15

Dmax%

0.10

0.09

0.11

0.06

Without (type-) recalibration the following Dmax values appear over the period of 4 days:

Dmax%

0.42

0.31

0.40

0.29

       Type recalibration largely eliminates instability, so it can be expected that AE spectrometers will be increasingly used for high-alloy qualities instead of XRF spectrometers. Besides being about 3 times cheaper, AES-SDAR allows analysis of C and  B as well as, more recently, gases eg. N in steel and O in a Cu base.











telephone:0086-15106177808
Tel:0510-85598302
email:lm@chxyq.com

Address: 19th Liangtong Road, Mashan Town, Binhu District, Wuxi City, Jiangsu Province, China

Copyright ? 2015 All Rights Reserved