Professional Documents
Culture Documents
Effect Coating
Effect Coating
ON
THE DURABILITY OF BANKNOTES
P. Koeze and A.H.B.Th. van Gelder
De Nederlandsche Bank N.V.
Amsterdam, Netherlands
May 1985
Digitised February 2007
Note
This paper was presented at the Madrid meeting of the Banknote Printers
Conference, 7 9 October 1986. This digital version was composed in February 2007 and follows the original text with minor emendations, with the
exception of two papers dating from the nineteen fties which were added
to the list of references.
Both authors worked in the Banknotes Technical Development Department of De Nederlandsche Bank N.V., Amsterdam, at the time this paper
was published. Dr Peter Koeze retired from the Bank in 2004; Ing Alwin
van Gelder is still employed by the Bank in 2007.
E-mail: [email protected]
List of Publications: http://www.pkoeze.nl/ListOfPublications.pdf
Curriculum vitae: http://www.pkoeze.nl/CurriculumVitae.pdf
De Nederlandsche Bank N.V.
P.O. Box 98
1000 AB Amsterdam
Netherlands
E-mail: [email protected]
Website: http://www.dnb.nl/
Summary
The eect of three dierent coating materials (UMC, CAP and Mowital
B3OH) on the durability of Netherlands f 25-banknotes has been investigated. Test series with the coatings were produced together with a test
series without coating and put into circulation. With the aid of the Banks
sorting system the gradual withdrawal of the banknotes has been continuously recorded for two-and-a-half years from the day of issue. It appears
that coated banknotes have a remarkably longer mean time to failure than
banknotes without coating. Of the coating materials tested CAP turns out
to be superior. On basis of this circulation trial, if CAP-coated banknotes
are used instead of uncoated banknotes, only 30% of the number of banknotes needs to be issued in order to keep the circulation at the same quality
level.
Introduction
297,000 0.9576
6805 6828 Mowital B3OH
1.30
294,300 0.7556
6829 6852 CAP
1.85
296,600 0.7146
6853 6876 UMC extra-thick 1.60
296,300 0.7256
The banknotes were issued in the period from Monday 19 April 1982
to Friday 30 April 1982. In the following evaluation Friday 23 April 1982
is taken as the date of issue. On Friday 30 November 1984, the last day
covered by the trial, the notes had been in circulation for 136 weeks and of
each variety 70% or more had been withdrawn.
For details of the method of calculation one is referred to the reports
mentioned [3] [4] [5].
2
2.1
Results
The return rate
The return rate o is dened as the ratio of the sum of t banknotes s and
unt banknotes v returned during a week t to the number of banknotes in
circulation c at the beginning of week t:
o (t) =
s (t) + v (t)
:
c (t 1)
(1)
Figure 3. t-statistic of three test series under the hypothesis that the return rate is constant. The 0.01 and 0.05 signicance levels are
indicated.
The conclusion is that the mean return rates of the test series are equal
and constant in the intervals mentioned. The exact values are given in table
3. Beyond those intervals the return rates decrease as a result of cullings
and rejects during sorting on the sorting machines [4] [5].
Table 3. Mean return rates of the test series and the entire f 25 circulation
coating
interval
mean return rate
(week)
(week 1 )
UMC
0 110
0.0742
UMC, extra-thick
0 110
0.0747
Mowital B3OH
0 110
0.0792
CAP
0 110
0.0705
no coating
0 66
0.0703
f 25 circulation (UMC) 0 110
0.0740
Finally, gure 4 shows the dierence between the return rates of the test
series and the entire f 25 circulation, giving further evidence that the return
rates of the test series and the entire circulation are equal in the intervals
mentioned.
Figure 4. Dierence between the return rates of the test series and of the
entire f 25 circulation.
O = UMC; X = UMC, extra-thick layer; H = Mowital B3OH;
Y = CAP; V = no coating; > = coinciding points.
2.2
The withdrawal rate q is dened as the ratio of the number of unt notes v
to the sum of t notes s and unt notes v returned during a week t:
q (t) =
v (t)
:
s (t) + v (t)
(2)
Figure 5 gives the withdrawal rates of the test series and gure 6 of the
entire f 25 circulation. Approximately 80% of the notes without coating
against only 25% of the coated notes are withdrawn in the asymptotic case.
10
11
2.3
12
13
M T T F1 =M T T F2 ) per cent.
(3)
29:9=99:4) = 70%:
In the case of Netherlands banknotes, which are coated with UMC already
since 1957, the change from UMC to CAP should reduce the consumption
with
100 (1 85:5=99:4) = 14%:
14
References
[1] N.E. Servatius; The last news about our one guilder notes treated with
Ultramid, Proceedings Banknote Printers Conference, Helsinki 1955.
[2] N.E. Servatius; Final conclusions about the series of one-guilder notes
treated with Ultramid, Proceedings Banknote PrintersConference, Vienna 1956.
[3] P. Koeze; An accurate statistical estimation of the life-length of f100banknotes. A circulation trial with two qualities of currency paper, Int.
Stat. Rev. 47 (1979) (3) 283297.
[4] P. Koeze; The life-length of banknotes. A case study, Statistica Neerlandica 36 (1982) 187207.
[5] P. Koeze and A.H.B.Th. van Gelder; The e ect of paper bre direction on
the durability of banknotes, De Nederlandsche Bank, Amsterdam 1983.
[6] M.B. Wilk, R. Gnanadesikan and M.J. Huyett; Estimation of parameters of the gamma distribution using order statistics, Biometrika 49
(1962) (3 and 4) 525545.
15