U.S. patent application number 11/569102 was filed with the patent office on 2007-12-27 for method for stabilizing the temperature dependency of light emission of an led.
This patent application is currently assigned to Target Systemelectronic GmbH. Invention is credited to Stein Jurgen, Guntram Pausch, Karen Saucke, Karen Saucke.
Application Number | 20070295912 11/569102 |
Document ID | / |
Family ID | 34957636 |
Filed Date | 2007-12-27 |
United States Patent
Application |
20070295912 |
Kind Code |
A1 |
Jurgen; Stein ; et
al. |
December 27, 2007 |
Method for Stabilizing the Temperature Dependency of Light Emission
of an Led
Abstract
Method for correction of the temperature dependency of a light
quantity L emitted by a light emitting diode (LED), being operated
in pulsed mode with substantially constant pulse duration t.sub.P,
and measured in a light detector, using a predetermined parameter
X, correlated to the temperature T of the LED in a predetermined
ratio, whereby a correction factor K is determined from the
parameter X, preferably using a calibration table, especially
preferred using an analytic predetermined function, whereby the
measured emitted light quantity L is corrected for the temperature
contingent fluctuations of the emitted light quantity, whereby the
parameter X is determined from at least two output signals of the
LED, which are related to each other in a predetermined manner.
Inventors: |
Jurgen; Stein; (Wuppertal,
DE) ; Pausch; Guntram; (Dresden, DE) ; Saucke;
Karen; (Solingen, DE) ; Pausch; Guntram;
(Dresden, DE) ; Saucke; Karen; (Sluttgart,
DE) |
Correspondence
Address: |
IP STRATEGIES
12 1/2 WALL STREET
SUITE I
ASHEVILLE
NC
28801
US
|
Assignee: |
Target Systemelectronic
GmbH
Koelner Strasse 99
Solingen
DE
42651
|
Family ID: |
34957636 |
Appl. No.: |
11/569102 |
Filed: |
May 14, 2004 |
PCT Filed: |
May 14, 2004 |
PCT NO: |
PCT/EP04/50813 |
371 Date: |
August 24, 2007 |
Current U.S.
Class: |
250/363.01 ;
250/216; 315/151 |
Current CPC
Class: |
G01J 1/02 20130101; H05B
45/18 20200101; Y02B 20/30 20130101; G01J 1/0295 20130101; H05B
45/12 20200101; G01J 1/30 20130101; G01T 1/40 20130101; H05B 31/50
20130101; G01J 1/08 20130101 |
Class at
Publication: |
250/363.01 ;
250/216; 315/151 |
International
Class: |
H05B 37/02 20060101
H05B037/02; G01T 1/20 20060101 G01T001/20 |
Claims
1. Method for correction of the temperature dependency of a light
quantity L emitted by a light emitting diode (LED), which is
operated in pulsed mode with substantially constant pulse duration
t.sub.P, and is measured in a light detector, using a predetermined
parameter X related to the temperature T of the LED in a
predetermined relation, whereby a correction factor K is determined
from the parameter X, preferably using a calibration table,
especially preferred using an analytic predetermined function,
according to which the measured emitted light quantity L is
corrected for the temperature contingent fluctuations of the
emitted light quantity, whereby the parameter X is determined from
at least two output signals of the LED, which are related to each
other in a predetermined manner.
2. Method for correction of the temperature dependency of the light
quantity L emitted by a light emitting diode (LED) according to
claim 1 and measured in a light detector, whereby at first the
temperature T of the LED is determined from the measured parameter
X, preferably using a calibration table, especially preferred using
an analytic predetermined function, and then, the correction factor
K is determined from the temperature T, preferably using a
calibration table, especially preferred using an analytic
predetermined function.
3. Method for temperature stabilization of a light emitting diode
(LED), which is operated in pulsed mode with substantially constant
pulse duration t.sub.P, whereby a predetermined parameter X,
related to the temperature T of the LED in a predetermined
relation, is used as a command variable, whereby the parameter X is
determined from at least two output signals of the LED, which are
related with respect to each other in a predetermined manner.
4. Method according to claim 1, whereby the parameter X is
determined as follows: Operating the LED in pulsed mode such that
the pulse duration tp remains substantially constant and the
voltage at the LED alternates periodically between at least a first
voltage U.sub.P1 and at least a second voltage U.sub.P2, Measuring
the average light quantities L(U.sub.P) of the pulses with at least
the voltages U.sub.P1 and U.sub.P2, Determining the parameter X
from the ratio of the light quantities.
5. Method according to claim 1, whereby the parameter X is
determined as follows: Operating the LED in pulsed mode such that
the pulse duration t.sub.P remains substantially constant and the
current flowing through the LED alternates periodically between at
least a first value I.sub.P1 and at least a second value I.sub.P2,
Measuring the average light quantities L(I.sub.P) of the pulses
with at least the currents I.sub.P1 and I.sub.P2, Determining the
parameter X from the ratio of the light quantities.
6. Method according to claim 1, whereby the parameter X is
determined as follows: Operating the LED in pulsed mode such that
the pulse duration t.sub.P takes substantially two different,
substantially constant, values t.sub.PS and t.sub.PL, and the
voltage at the LED alternates periodically between at least a first
voltage U.sub.P1 and at least a second voltage U.sub.P2, Measuring
the average light quantities L(U.sub.P; t.sub.PS) and L(U.sub.P;
t.sub.PL) of the pulses with at least the voltages U.sub.P1 and
U.sub.P2 and the pulse durations t.sub.PS and t.sub.PL, Determining
at least the differences D.sub.P1 and D.sub.P2 of the light
quantities L(U.sub.P1; t.sub.PL) and L(U.sub.P1; t.sub.PS), as well
as L(U.sub.P2; t.sub.PL) and L(U.sub.P2; t.sub.PS), and Determining
the parameter X from the ratio of the differences of the light
quantities.
7. Method according to claim 1, whereby the parameter X is
determined as follows: Operating the LED in pulsed mode such that
the pulse duration t.sub.P takes substantially two different
substantially constant values t.sub.PS and t.sub.PL, and the
current flowing through the LED alternates periodically between at
least a first value I.sub.P1 and at least a second value I.sub.P2,
Measuring the average light quantities L(I.sub.P; t.sub.PS) and
L(I.sub.P; t.sub.PL) of the pulses with at least the current
I.sub.P1 and I.sub.P2 and the pulse durations t.sub.PS and
t.sub.PL, Determining at least the differences D.sub.P1 and
D.sub.P2 of the light quantities L(I.sub.P1; t.sub.PL) and
L(I.sub.P1; t.sub.PS) as well as L(I.sub.P2; t.sub.PL) and
L(I.sub.P2; t.sub.PS), and Determining the parameter X from the
ratio of the differences of the light quantities.
8. Method according to claim 4, whereby the light quantities
L(U.sub.P) and L(I.sub.P) are determined with a light detector,
preferably a photo multiplier, a hybrid photo multiplier, an
Avalanche photo diode or a photo diode with amplifier.
9. Method according to claim 8, whereby the light quantities
measured by the light detector are determined by one or more of the
following process steps: Carrying out the pulse amplitude
spectrometry of the detector signals, Measuring the average current
flow in the light detector, Measuring the charge quantity of the
LED pulse generated in the photo sensitive layer of the light
detector, preferably by means of spectrometry of the preferably
already amplified charge signals triggered by the LED pulse.
10. Method according to claim 1, whereby the LED comprises a series
resistance.
11. Method according to claim 10, whereby the series resistance is
selected such that its resistance depends on the temperature T in a
non-linear manner, preferably in a manner such that the dependency
or at least the non-linearity of the dependency of the correction
factor K from the temperature T is compensated for approximately by
the temperature dependency of the series resistance.
12. Method for stabilization of a light detector, preferably a
photo multiplier, a hybrid photo multiplier, an Avalanche photo
diode or a photo diode with amplifier, whereby the light detector
is optically connected with at least an LED, whereby a least one
LED is operated in pulsed mode, and whereby the output signals of
the light detector are stabilized with a stabilizing factor,
whereby the stabilizing factor is generated by the signals emitted
by the at least one LED, and whereby the temperature dependency of
the light emission of at least one LED is corrected with a method
according to claim 1.
13. Method for stabilization of the signals generated by a
scintillation detector for measuring radiation, preferably ionized
radiation, by the radiation absorbed at least partially in the
detector, and depending on the operating temperature of the
detector, whereby the scintillation detector has at least one light
detector and at least one LED optically connected to the latter,
whereby the stabilization factor for stabilizing the scintillation
detector is generated from the signals emitted by the at least one
LED, and whereby the temperature dependency of the light emission
of at least one LED is corrected by means of a method according to
claim 1.
14. Method according to claim 13, whereby at least one LED is
connected to the scintillator in a heat conducting manner.
15. Method according to claim 1, whereby the signal processing is
carried out digitally.
16. Light detector, preferably a photo multiplier, a hybrid photo
multiplier, an Avalanche photo diode or a photo diode with
amplifier, with signal processing device, whereby at least an LED
is connected to the light detector optically, whereby at least one
LED is operated in pulsed mode, and the output signals of the light
detector are stabilized with a stabilizing factor, whereby the
stabilizing factor is generated by signals emitted by the at least
one LED, and whereby the temperature dependency of the light
emission of at least one LED is corrected by means of a method
according to claim 1.
17. Light detector according to claim 16, whereby the signal
processing is carried out digitally.
18. Scintillation detector for measuring radiation, preferably
ionized radiation, whereby the scintillation detector has a light
detector according to claim 16, which measures light generated by
the scintillation detector at least partially.
19. Scintillation detector according to claim 18, whereby the
signals generated by the radiation absorbed in the detector at
least partially and depending on the operating temperature of the
detector are measured and are stabilized with a stabilizing factor
S related to the temperature T of the scintillator in a
predetermined ratio, whereby at least one LED of the light detector
is connected to the scintillation detector in a heat conducting
manner, and whereby the temperature contingent stabilizing factor S
for stabilizing the scintillation detector is determined in a
predetermined manner, preferably using a calibration table, in
particular preferred using a predetermined functional dependency
from the parameter X of at least one LED connected to the
scintillation detector in a heat conducting manner, according to a
method for correction of the temperature dependency of a light
quantity L emitted by a light emitting diode (LED), which is
operated in pulsed mode with substantially constant pulse duration
t.sub.P, and is measured in a light detector, using a predetermined
parameter X related to the temperature T of the LED in a
predetermined relation whereby a correction factor K is determined
from the parameter X, preferably using a calibration table
especially preferred using an analytic predetermined function
according to which the measured emitted light quantity L is
corrected for the temperature contingent fluctuations of the
emitted light quantity, whereby the parameter X is determined from
at least two output signals of the LED, which are related to each
other in a predetermined manner.
Description
[0001] The invention relates to a method for correcting the
temperature dependency of a light quantity emitted by a
light-emitting diode (LED), which is operated in pulsed mode with
substantially constant pulse duration, and measured in a light
detector.
[0002] The emitted light quantity of an LED depends on its
temperature. In laboratory applications according to prior art,
according to which an LED is employed as reference light source,
the LED and possibly the measuring apparatus associated therewith
are tempered, resulting in the temperature and, thus, the emitted
light quantity of the LED remaining constant.
[0003] In applications outside the laboratory, in which such a
climatisation is not possible at all or only at increased
expenditure, it is therefore necessary to correct the measured
values of the light quantity with respect to the temperature
contingent influences, to thereby reduce the errors of the measured
result. In case such an LED is used for example as light source for
stabilization of a photo multiplier, which for example is employed
as light detector in a scintillation detector, for example a mobile
detector for identification of radio isotopes (hand held radio
isotope identification device--RID), the LED is exposed to thermal
fluctuations in the range of -20.degree. C. to +50.degree. C.
Thereby, the system amplification of the light detector can
fluctuate offhand for about 20% and more, such that a stabilization
of the amplification of the light detector is necessary, to
maintain the energy amplification and the energy resolution of the
RID sufficiently good. For stabilization of Such a light detector
with an LED, it is therefore necessary, to know the temperature
dependency of the light quantity emitted by the LED.
[0004] Methods for stabilization are known, according to which the
temperature is measured at or in the detector and the temperature
caused effects are adjusted by means of previously measured
calibration tables. These methods, however, have the drawback that
a temperature measurement with fast temperature changes is only
hardly realizable, particularly for the reason that often no
uniform temperature distribution can be expected in the detector.
Besides, the amplification of, for example, a photo multiplier does
not only depend on its temperature, but rather also on the
effective counting rate and its previous history, i.e. its
hysteresis and age. It has been found that sufficiently exact
prediction of the amplification under consideration of all
parameters is not possible.
[0005] For stabilization, therefore, often active methods are
employed during the actual measurement. Mostly, radio active
calibration sources or natural background radiation are used, to
achieve such an active stabilization. This, however, leads to
optimization problems, because a compromise of sufficiently short
but nevertheless sufficiently exact calibration measurements has to
be found. Additionally, each additional radio active radiation
leads to a reduction of the total sensitivity of the system.
[0006] An alternative is the separated stabilization of light
detector and scintillator--the latter is for example disclosed in
PCT/EP2004/050754. It is known to use a pulsed light source, for
example an LED, as measured standard for the stabilization of the
light detector. It is also known to stabilize and to monitor the
amplification of light detectors in this manner in laboratory
applications. Disadvantageous with respect to this prior art is
that the light emission of an LED depends on its temperature, more
particular, on its junction temperature T.sub.LED. Thus, according
to known methods, it is either necessary, to keep the temperature
constant or to monitor it at least, or to monitor the light
quantity emitted respectively by the LED with a separate
measurement apparatus precisely. Such an assembly is not only
technically complex and cost intensive, but rather requires also
additional energy and additional space, complicating the use in
battery operated mobile RIDs.
[0007] From sensor techniques, a method is known, to measure the
temperature of semiconductor elements by means of a current
measurement at constant operating voltage or by means of a
measurement of the flux voltage at constant current.
[0008] Therefore, it is an object of the invention, to provide a
method avoiding the drawbacks of prior art mentioned above, to
reduce the expenditure for the stabilization of light detectors by
means of pulsed LEDs.
[0009] Further, it is an object of the invention, to provide a
light detector, the signals of is which, including the pulse
amplitude spectrum produced by the associated electronics, can be
corrected and, thus, stabilized by means of pulsed LED with respect
to temperature dependency and otherwise caused fluctuations.
Moreover, it is an object of the invention, to provide a detector
for measuring radiation, preferably ionized radiation, which can be
stabilized by a pulsed LED.
[0010] These objects are at first solved by the method and devices
according to the claims. Thus, a method is provided, according to
which the emitted light quantity L of a light emitting diode being
temperature dependent is corrected, using a predetermined parameter
X being in a predetermined relation to the temperature T of the
LED. From the parameter X, a correction factor K is thereby
determined, preferably using a calibration table, especially
preferred using an analytic predetermined function, according to
which the measured emitted light quantity L is corrected for the
temperature-caused fluctuations of the emitted light quantity.
Thereby, the diode is operated in pulsed mode with substantially
constant pulse duration t.sub.P. The parameter X, thereby, is
determined from at least two output signals of the LED itself,
which are related with respect to each other in a predetermined
manner.
[0011] Thereby, it has been found to be advantageous, to determine
at first the temperature T of the LED from the measured parameter
X, whereby a calibration table can be used. Preferably, it is also
possible to use an analytic predetermined function. Subsequently,
the correction factor K is determined from the temperature T,
whereby also preferably a calibration table or an analytic
predetermined function is used.
[0012] Moreover, a method for temperature stabilization of a light
emitting diode (LED) is provided, whereby the LED is operated in
pulsed mode with substantially constant pulse duration t.sub.P,
whereby a predetermined parameter X is used as command variable,
associated to the temperature T of the LED in a predetermined
relation, whereby the parameter X is determined from at least two
output signals of the LED, which are related to each other in a
predetermined manner.
[0013] It has been found to be advantageous, to operate the LED
such that the pulse duration t.sub.P is substantially constant, the
voltage applied to the LED, however, changing between at least one
first voltage U.sub.P1 and at least a second voltage U.sub.P2,
being different from the first voltage U.sub.P1. During the pulse,
the respective voltage is substantially constant. Then, the average
light quantities L(U.sub.P) of the pulses at different voltages
U.sub.P are measured, thus, at least the average light quantity
L(U.sub.P1) of the pulse at voltage U.sub.P1 and the average light
quantity L(U.sub.P2) of the pulse at voltage U.sub.P2. The
determination of the parameter X is then derived from the ratio of
the light quantities L(U.sub.P) with respect to each other. The use
of the ratio of at least two light quantities at constant pulse
duration but at different voltages leads to the fact that
amplification fluctuations of the light detector caused by
temperature fluctuations or by other effects do not have any
influence on the determination of the parameter X.
[0014] The method can also be configured such that a current to the
LED being in pulsed mode at also constant pulse duration t.sub.P,
periodically alternating between at least a first value I.sub.P1 or
at least a second value I.sub.P2, being different from the first
one, is applied. During the pulse, the current, flowing through the
LED, is respectively substantially constant. Then, the average
light quantities L(I.sub.P) of the pulses with the different
currents I.sub.P, thus, at least the average light quantity
L(I.sub.P1) of the pulse with the current I.sub.P1 and the average
light quantity L(I.sub.P2) of the pulse with the current I.sub.P2,
are measured. The parameter X is then determined from the ratio of
the light quantities L(I.sub.P) with respect to each other.
[0015] To suppress the influence of turn on and turn off effects or
similar influences to the light emission of the LED, it has been
found to be especially advantageous, to determine the parameter X
as follows: Operating the LED in pulsed mode such that the pulse
duration t.sub.P takes substantially two different substantially
constant values t.sub.PS and t.sub.PL and the voltage alternates
between at least a first voltage U.sub.P1 and at least a second
voltage U.sub.P2, being different from the first voltage U.sub.P1
periodically at the LED, measuring the average light quantities
L(U.sub.P; t.sub.PS) and L(U.sub.P; t.sub.PL) of the pulses with at
least the voltages U.sub.P1 and U.sub.P2 and the pulse durations
t.sub.PS and t.sub.PL, determining the differences D.sub.P1 and
D.sub.P2 of the light quantities L(U.sub.P1; t.sub.PL) and
L(U.sub.P1; t.sub.PS) as well as L(UP.sub.2; t.sub.P1) and
L(U.sub.P2; t.sub.PS), and determining the parameter X from the
ratio of the differences of the light quantities.
[0016] Just as well, it is possible to determine the parameter X as
follows: Operating the LED in pulsed mode such that the pulse
duration tp takes substantially two different substantially
constant values t.sub.PS and t.sub.PL, and the current flowing
through the LED alternates periodically between at least a first
value I.sub.P1 and at least a second value I.sub.P2, being
different from I.sub.P1, measuring the average light quantities
L(I.sub.P; t.sub.PS) and L(I.sub.P; t.sub.PL) of the pulses with at
least the currents I.sub.P1 and I.sub.P2 and the pulse durations
t.sub.PS and t.sub.P1, determining the differences D.sub.P1, and
D.sub.P2 of the light quantities L(I.sub.P1; t.sub.PL) and
L(I.sub.P1; t.sub.PS) as well as L(I.sub.P2; t.sub.PL) and
L(I.sub.P2; t.sub.PS), and determining the parameter X from the
ratio of the differences of the light quantities.
[0017] Further, it has been found to be advantageous, if the light
quantities L(U.sub.P;) and L(I.sub.P), respectively, i.e. at least
the light quantities L(U.sub.P1) and L(U.sub.P2) or L(I.sub.P1) and
L(I.sub.P2), are determined with a light detector, preferably a
photo multiplier, a hybrid photo multiplier, an Avalanche photo
diode or a photo diode with amplifier. The light quantities
measured with this light detector are preferably determined by
application of one or more of the following method steps: Carrying
out pulse amplitude spectrometry of the detector signals and/or
measuring the average current flow in the light detector and/or
measuring the charge quantity produced in the photo sensitive layer
of the light detector by the LED pulse, preferably by means of
spectrometry of the, already amplified, charge signals triggered by
the LED pulses.
[0018] It is further advantageous, if the LED comprises a series
resistance, whereby the series resistance is selected particularly
advantageous in that its resistance does not depend on temperature
T in a linear manner, especially preferred in a manner that the
dependency or at least the non-linearity of the dependency of the
correction factor K from the temperature T is compensated
approximately by the temperature dependency of the series
resistance.
[0019] Further, a method for stabilizing a light detector is
claimed, preferably a photo multiplier, a hybrid photo multiplier,
an Avalanche photo diode or a photo diode with amplifier, whereby
the light detector is optically connected to at least an LED,
whereby at least an LED is operated in pulsed mode and according to
which the output signals of the light detector are stabilized with
a stabilizing factor, whereby the stabilizing factor is generated
by the signals of the at least one LED and according to which the
temperature dependency of the light emission of at least one LED is
corrected by means of one of the methods described above.
[0020] Further, a method for stabilization of signals generated by
a scintillating detector for measuring radiation is claimed,
preferably ionized radiation, whereby the signals are generated by
the radiation which is at least partly absorbed in the detector,
and which depend on the operating temperature of the detector,
whereby by scintillating detector has at least one light detector
and at least one LED optically connected thereto, whereby the
stabilizing factor for stabilizing the scintillation detector is
generated from the signals emitted by at least one LED, and
according to which the temperature dependency of the light emission
of the LED is corrected according to one of the methods described
above and claimed in claims 1 to 11. It can also be an advantage,
if at least one, preferably the optical connection between the LED
and the scintillating detector is designed in a heat conducting
manner, because then the temperature of the LED being
heat-conducingly connected to the scintillating detector
substantially corresponds to the temperature of the
scintillator.
[0021] In all the methods described above, signal processing is
preferably carried out digitally.
[0022] Moreover, a light detector with a signal processing device
is claimed, preferably a photo multiplier, a hybrid photo
multiplier, an Avalanche photo diode or a photo diode with
amplifier, whereby at least one LED is optically connected to the
light detector, according to which at least an LED is operated in
pulsed mode and the output signals of the light detector are
stabilized by a stabilizing factor, whereby the stabilizing factor
is generated from the signals generated by the at least one LED,
and according to which the temperature dependency of the light
emission of at least one LED is corrected with a method described
above and claimed in claims 1 to 11. Here, the signal processing
preferably is carried out digitally.
[0023] Further, a scintillation detector for measuring of radiation
is claimed, preferably ionized radiation, whereby the scintillation
detector has at least one light detector described above, measuring
the light generated by the scintillation detector at least
partially. In a special embodiment, signals are measured which are
generated by the radiation absorbed at least partially in the
detector and being dependent on the operating temperature of the
detector, and are stabilized by a stabilizing factor being in a
predetermined relation to the temperature T of the scintillator,
whereby at least an LED of the light detector is connected to the
scintillation detector in a heat conducting manner, and whereby the
temperature dependency stabilizing factor S for stabilizing the
scintillation detector in a predetermined manner, preferably using
a calibration table, in particular preferred using a predetermined
functional dependency, is determined from parameter X of at least
one LED being connected to the scintillation detector in a
heat-conducting manner according to one of the process steps
described above.
[0024] The present invention provides a technically very simple and
convenient method for temperature stabilization of LED reference
light sources, which, for example, are used for stabilization of
light and/or scintillation detectors, in that it analyses the pulse
amplitude spectrum of LED signals, which have to be measured anyway
for stabilization. Therefore, neither a radio active calibration
source is necessary, nor the use of an additional light detector
for monitoring the light quantity emitted by the LED in dependency
from the temperature. The light detector, being present anyway, is
sufficient, the stability of which does not matter anyway, as long
as its amplification only alternates in periods of time, which are
larger than the switching interval of the different LED modi. This
switching interval can be kept very small (up to <1 ms), but is
at least as large as the temporal distance between two LED
pulses.
[0025] In the following, preferred embodiments are described by
means of the figures, described subsequently, showing:
[0026] FIG. 1 Measuring device for calibrating the LED;
[0027] FIG. 2 Voltage courses U.sub.P1 and U.sub.P2 at the LED
depending on time t;
[0028] FIG. 3a Voltage courses for voltages U.sub.P at different
pulse lengths t.sub.PS and t.sub.PL depending on time t;
[0029] FIG. 3b Schematic illustration of the difference D.sub.P
being derived from two signals of different length at equal voltage
U;
[0030] FIG. 4 Pulse amplitude spectrum of LED pulses;
[0031] FIG. 5 Dependency of the light quantity L and the correction
factors K of the LED temperature of two operating regimes of the
LED (measured values), derived therefrom;
[0032] FIG. 6 Dependency of the light quantity ratio R, determined
from the measured values shown in FIG. 5 from the LED
temperature;
[0033] FIG. 7 Dependency of the light quantity L and the correction
factors K of the light quantity ratio R (measured values) derived
therefrom;
[0034] FIG. 8 Dependency of the ratio from the light quantity
differences D.sub.P1 and D.sub.P2 from the LED temperature.
[0035] In a test and calibration device according to FIG. 1, an LED
is arranged such that it can be heated and cooled independent of
the scintillation detector, comprising an NaI(T.sub.1)-scintillator
crystal and photo multiplier PMT, by means of a Peltier element P
and a cooling element H. Thereby, the LED is accommodated in a
tempered aluminum block, the temperature of which is measured with
a conventional temperature sensor T. The light of the LED enters
into the scintillator crystal through an optical window O.
[0036] The LED is supplied with an adjustable voltage by a driver
circuit in pulsed mode. The driver circuit itself, as also the
entire electronics otherwise required, is shown in FIG. 1 only
schematically. The pulse length is also adjustable and is
stabilized by a quartz generator. It can be varied controlled by a
program during the measurements between several fixed values.
[0037] A further control unit provides for switching the voltage
applied during the pulse to the diode regularly automatically in a
time lag of respectively several seconds between two
pre-determinable stabilized values U.sub.P1 and U.sub.P2.
[0038] The shapes of the pulses applied to the diode respectively
are illustrated schematically in FIGS. 2, 3a and 3b, which are
further explained below.
[0039] For checking the function of the system, a radioactive
.sup.137Cs-source is fixed to the NaI(T1) scintillator, which
generates corresponding signals in the scintillation detector. The
entire arrangement, including electronics, is accommodated in a
climatic cabinet, the interior temperature of which can be modified
controlled by a program or can be kept constant.
[0040] For the measurements described in the following, an LED
having a maximum of the wave length spectrum in the blue range,
namely at approximately 430 nm, was used as an example,
corresponding approximately to the spectral distribution of the
emission light of an NaI(T1) scintillator crystal.
[0041] It is known that the average light quantity emitted by the
LED depends on its temperature at otherwise constant operating
conditions. The corresponding temperature dependency was measured
by means of this arrangement as follows: [0042] keeping climatic
cabinet at constant temperature [0043] increasing or reducing
temperature T.sub.LED of the LED by means of the Peltier element
[0044] measuring temperature T.sub.LED of the LED by means of a
temperature sensor until T.sub.LED remains constant [0045]
recording and analyzing the pulse amplitude spectrum in the
selected measuring regime.
[0046] FIG. 4 shows exemplaryly a measured pulse amplitude
spectrum. During the measurement, U.sub.P was switched regularly
between the values U.sub.P1 and U.sub.P2, the pulse length t.sub.P
was regularly switched between the values t.sub.PS and t.sub.PL.
Each combination of U.sub.P and .sub.P generates a peak in the
spectrum, the position of which is a measure for the light quantity
L(U.sub.P, t.sub.P) emitted by the diode on average and detected in
the detector. In the left part of the drawing, the pulse amplitude
spectrum generated by the .sup.137Cs source in the scintillating
detector can be seen additionally. In the following drawings and
formulas, L is equated to the position of the corresponding peak
due to the fixed relationship of the two variables with respect to
each other.
[0047] FIG. 5 shows the measured dependency of the emitted light
quantity L from the temperature T.sub.LED for two different
operating regimes, which are characterized by equal pulse lengths,
but different voltage values U.sub.P1 and U.sub.P2. The variation
of the light quantity L with respect to the temperature T.sub.LED
can be clearly seen. Also shown are the correction factors
K(U.sub.P1)=L.sub.0/L.sub.1(T.sub.LED) and
K(U.sub.P2)=L.sub.0/L.sub.2(T.sub.LED) by means of which the light
quantities L and, thus, the peak positions of the peak measured
with respect to the pulses with the voltages U.sub.P1 and U.sub.P2
have to be corrected, to correct temperature contingent changes of
the light emission of the LED.
[0048] The corresponding voltage pulses respectively applied to the
LED are shown in FIG. 2 schematically, in which the two voltages
U.sub.P1 and U.sub.P2 as well as the pulse length t.sub.P are
shown. There, also the finite signal rising and falling times as
well as the transient effects during turning on of the signals are
shown.
[0049] From the two peak positions, the ratio
R=L(U.sub.P1)/L(U.sub.P2) is determinable. Due to the
non-linearities of the characteristic curve of the LED, this ratio
is not constant, but rather changes as shown in FIG. 6 with the
temperature T.sub.LED. The variable R, thus, determined, does not
depend on the amplification of the light detector, as long as the
detector signals are proportional to the light quantity L detected.
This proportionality (linearity of the detector response) is
actually given and was detected in additional measurements.
[0050] Thus, R can be measured with an un-stabilized detector with
unknown amplification. From R, then the temperature T.sub.LED can
be determined by means of the calibration curve (FIG. 6) of the
diodes.
[0051] From T.sub.LED, a temperature dependent factor
K=L.sub.0/L(T.sub.LED) can be determined with the dependencies
L(T.sub.LED) for each operating regime, which can correct the
absolute light quantity L measured at the detector for temperature
contingent fluctuations (FIG. 5). L.sub.0 hereby is the
corresponding peak position at a reference temperature T.sub.0. The
LED becomes measurement standard for the stabilization of the
amplification of the light detector due to the knowledge of the
correction factor K in spite of a change of the temperature
T.sub.LED. The amplification of the light detector (photo
multiplier) can either be controlled by means of these values or
can be corrected mathematically such that the currently measured
LED peak position, for example for U.sub.P1 and t.sub.P1, corrected
for the corresponding correction factor K, are shifted to the
desired position. Thereby, it is ensured that the detector signal
generated by a particular defined light quantity always generates a
peak at the same position in the spectrum--the amplification of the
light detector is stabilized.
[0052] Thus, R is an adequate parameter X within the above
mentioned meaning, which is derived from signals of the light
detector, corresponding to different operating modes of the LED,
and allows a determination of the temperature of the LED or a
correction of the temperature dependency of the emitted light
quantity.
[0053] In FIG. 7, the absolute peak position L and the
corresponding correction factors K for two different operating
regimes are illustrated as a function of the--respectively
determined from the same pulse amplitude spectrum--parameter R. The
drawing shows that the determination of the temperature T.sub.LED
can be omitted and instead, the decisive correction factor K is
correlated directly to the ratio R, and then can be respectively
determined from the currently measured R.
[0054] In the arrangement described above, the LED is operated with
pulses of constant voltage. It is also possible, to operate the
diode with pulses of constant current strength, and then to measure
the resulting peak positions. The variable I.sub.P, and U.sub.P are
linked to each other unambiguously via the diode characteristic
line of the respectively used LED. Although, the corresponding
dependencies of the peak positions L and the peak position ratios R
from the temperature T.sub.LED have another shape, they can,
however, be used in the same manner for determination of a
correction K correcting the temperature dependency of the light
emission L of the diode.
[0055] Although the formation of the light quantity ratios already
arranges for the amplification drifts of the light detector not
having an influence on the determination of the correction factor
R, turning on and turning off procedures can influence the light
emission of the LED and their temperature dependency in an
undesired manner. These effects can be additionally reduced with a
further embodiment, being based on the measurement of more than two
pulses.
[0056] In the following, an embodiment is described in detail,
according to which as shown in FIG. 4, four pulses are measured,
and, in fact, respectively two pulses at constant voltage U.sub.P1
and different pulse durations t.sub.PS and t.sub.PL as well as two
pulses of also constant voltage U.sub.P2 and also again different
pulse durations t.sub.PS and t.sub.PL. However, also a larger
number of pulses can be measured and analyzed, to increase the
precision further.
[0057] FIG. 3a shows schematically the voltage course of two pulses
with different pulse durations t.sub.PS and t.sub.PL at otherwise
identical edge conditions, in particular at identical diode voltage
U.sub.P1. Because the voltages U.sub.P1 of the two pulses are
identical, the setting time as well as the rise time of the pulses
are substantially identical also. If now the light quantity
differences L.sub.D=L(U.sub.P1; t.sub.PL)-L(U.sub.P1; t.sub.PS) of
the two pulses at constant edge conditions but different pulse
durations are established, the identical diode-typical parameters
are subtracted from each other and suppressed as a result such that
the light quantity difference corresponds substantially to an area
of the plateau region of the pulse, and therefore is determinable
exactly. This is shown schematically in FIG. 3b.
[0058] This light quantity difference is now established for two
different voltages U.sub.P1 and UP.sub.2 in the described manner,
and subsequently the light quantity ratio of L.sub.D1 to L.sub.D2
is determined. This light quantity ratio is especially adequate to
serve as parameter X, from which the correction factor K is
determined: K = f .function. ( X ) ##EQU1## with ##EQU1.2## X = L D
.times. .times. 1 L D .times. .times. 2 = L .function. ( U P
.times. .times. 1 ; t PL ) - L .function. ( U P .times. .times. 1 ;
t PS ) L .function. ( U P .times. .times. 2 ; t PL ) - L .function.
( U P .times. .times. 2 ; t PS ) ##EQU1.3##
[0059] The result of measurements of the ratio of the light
quantity differences for two different voltages and pulse lengths
shows that this parameter correlates very well with the temperature
T.sub.LED (FIG. 8).
[0060] Naturally, here also the variation of the light quantity
cannot only result via the voltage U.sub.P, but rather also via a
variation of the diode current I.sub.P.
[0061] For measuring the light quantities, in the present example a
photo multiplier was used, whereby it is just as well possible, to
use a photo diode or another form of the light detector.
[0062] As FIG. 6 and FIG. 8 show, the dependency of the parameter X
from the diode temperature T usually is non-linear. The electronic
determination of the correction factor K can be further simplified,
if the LED is operated with a series resistance, and preferably
this series resistance is selected such that this also has a
non-linear behavior with respect to temperature changes. In case
the parameter hereby is selected such that the non-linearity of the
series resistance in the respectively interesting temperature range
approximately compensates for the non-linearity of the parameter X,
then eventually an approximately linear dependency of the
correction factor K from the diode temperature T is obtained in the
interesting temperature range, substantially simplifying the
analysis and correction of the measurement results.
[0063] In practical application, the corresponding dependencies are
to be determined previously, i.e. are to be measured usually such
that a characteristic curve is generated for the concretely
employed LED. This, then can be stored in form of a correction
table or also of an analytic function, such that a correction of
the measurement even in real time during the measurement itself can
result.
[0064] For a sufficient stabilization it is satisfying, if the
parameters X and K are determined in intervals, which are smaller
than the time periods, in which a relevant temperature change of
the LED takes place. Thereby, it is obvious that the temperature
T.sub.LED of the LED does not have to be explicitly known for the
correction of the measured light quantity, because the correction
results from the signals themselves. Anyhow, it is clear that the
temperature T.sub.LED of the LED can also be determined by means of
this method, whereby the determination of the correction factor K
then is determined as variable also derived from the diode
temperature T.sub.LED. In such cases, the measurement of the diode
temperature T.sub.LED can be used for other purposes as well.
[0065] Such temperature measurements with the LED can, for example,
be used for the calibration of a scintillation detector. Such a
scintillation detector usually consists of a scintillator in solid,
crystal or liquid form, as well as a light detector. The
characteristics of the light detector, in most cases a photo
cathode with photo multiplier or photo diode, depend on the
temperature of photo cathode and in particular, photo multiplier.
In case, a constant light quantity L is radiated into the light
detector, the output signal of the light detector can be
calibrated, to thus compensate for temperature contingent
fluctuations. Thereby, it is not necessarily required that the
radiated light quantity remains constant, but it is rather
sufficient, if this is known.
[0066] In case an LED is used, connected at or in the scintillator
or otherwise in the range of sight of the light detector, the
emitted light quantity of which is temperature corrected according
to one of the methods described above, and thus is known, the
entire light detector can be calibrated with such an LED, whereby
this can result online during the measurement. Due to the reasons
mentioned above it is, however, possible to measure the temperature
T of the system, but, in fact, it is not necessary, because the
analysis of the signal value is sufficient for the calibration.
[0067] In case the LED used including the light detector is coupled
to the scintillator in such a manner that the temperature of the
LED corresponds substantially to that of the scintillator, it is
moreover possible, to achieve a stabilization of the entire system
against temperature dependency changes of the system amplification
by means of the LED.
[0068] It is known that also the light emission of a scintillator
depends on the temperature T. Is the temperature of the
scintillator known, possibly by the analyzing of signals sent from
the LED according to one of the methods described above, the yield
of light of the scintillator depending on the temperature as well
as the dependency of the light detector on the temperature during
the measurement, can be considered such that calibration of the
total system is possible over the analysis of the signals sent from
the LED. Because the yield of light of the scintillator corresponds
otherwise substantially to the energy of the radiation absorbed
there, therewith an energy calibration of the entire detector
system can result over the signal analysis of the signals triggered
by the LED at the end of the light detector, without the use of,
for example, a radio active calibration source being necessary for
calibrating the detector system.
* * * * *