EVALUATION OF THE MEASURING SIGNAL OF A VACUUM LEAK DETECTOR

Description

[0001] The invention relates to a method and a device for evaluating the measurement signal of a vacuum leak detector.

[0002] Vacuum leak detectors are used to detect leaks in test objects. For this purpose, a vacuum leak detector comprises a test chamber for the test object, a vacuum pump connected to the test chamber for evacuating the test chamber, and a gas detector connected to the test chamber, which is designed to analyze the gas extracted from the test chamber. In particular, the gas detector is designed to detect a test gas present in the test object that enters the test chamber through a leak in the test object and is then fed to the gas detector.

[0003] The pumping speed of the vacuum leak detector for the test gas used, usually helium, is determined by the vacuum pump used as well as the cross-section and length of the connection to the test chamber, i.e. the volume of the gas line between the test chamber and the gas detector. This effective test gas pumping speed, in relation to the volume of the test object, determines the system time constant of the vacuum leak detector. The system time constant is the duration of the system-related increase in the measured signal in response to a gas component at the gas sensor or the duration of the system-related decrease in the measured signal in response to a gas component that is no longer present at the gas sensor. The gas component at the gas sensor is typically the test gas component that has escaped through a leak and is to be detected for leak detection. Typically, the system time constant is assumed to be the time until the measurement signal drops to its 1 / e-th fraction orThe time required for the measurement signal to rise to the (1-1 / e)th fraction when a leak is sprayed with test gas or opened, or when a leak is no longer sprayed or closed. For simplicity, the vacuum time constant of the vacuum leak detector can also be assumed as the system time constant. The vacuum time constant is equal to the volume of the test object or test chamber divided by the effective pumping speed of the vacuum leak detector.

[0004] The pumping speed can be increased by using a larger pump, increasing the volume, or increasing the cross-section, thereby improving the performance and response time of the vacuum leak detector. However, the volume, cross-section, and size of the pump cannot be increased indefinitely. Furthermore, the resulting amplitude of the leak signal is reduced at large volumes, making leak detection more difficult.

[0005] The invention is based on the object of providing a method and a device which enable improved signal evaluation with a shortened reaction time.

[0006] DE 11 2014 004 741 T5 discloses a leakage testing device.

[0007] EP 2 686 657 A1 discloses a method and an arrangement for detecting leaks.

[0008] DE 199 42 185 A1 discloses a device for determining leakage currents.

[0009] EP 3 499 206 A1 describes a leak testing system for testing the leak tightness of a container.

[0010] The method according to the invention is defined by the features of patent claim 1. The vacuum leak detector according to the invention is defined by the features of patent claim 16.

[0011] According to the method according to the invention, the system time constant τ of the vacuum leak detector is first determined. The system time constant τ represents the system's response time to a change in the gas conditions at the gas detector. The system time constant τ thus corresponds to the duration of the system-related increase in the measured signal in response to the detection of a gas component escaping through a leak in the test object and / or the duration of the system-related decrease in the measured signal in response to a gas component from a gas leak no longer being detected. In the simplest case, the system time constant τ corresponds, for example, to the time that elapses until the measurement signal has decayed to a 1 / e-th fraction of the original signal, where e is Euler's number.

[0012] After the system time constant has been determined, the actual leak detection is performed by placing a test specimen into the test chamber and evacuating the chamber using the vacuum pump. A measurement signal I(t) of the gas extracted from the test chamber is generated by the gas detector at a current time t. The current time t can, for example, be the start of a new measurement or a new series of measurements, whereby measurements can be taken at regular intervals t0 within a series of measurements.

[0013] From the measurement signal I(t), a measurement signal Í(t+t0) predicted for a future time t+t0 is determined, assuming a decaying signal that follows exclusively the system time constant, typically an exponential decay. This means that to determine the predicted signal, it is assumed that the measurement signal I is subject to a purely exponential change and changes corresponding to the determined system time constant. This means that at time t+t0, the gas components are essentially present at the detector as expected exponentially from time t in vacuum technology, because no test gas flows into the test volume through a leak. So that the predicted measurement signal Í(t+t0) is determined solely by taking into account the signal curve during the system time constant τ. The system time constant τ is typically a few seconds or a few tens of seconds, for example, approximately 30 seconds.The time t+t0 lies at a maximum of the system time constant τ in the future from time t, i.e., τ > t0. Preferably, t0 is an nth part of τ, where n is a natural number. Typically, t0 is in the range of a few seconds, especially in the case of a measurement series with recurring measurements at regular intervals t0.

[0014] The predicted measurement signal Í(t+t0) therefore indicates the value to which the measurement signal I(t) will have risen or fallen after the time t0 < τ has elapsed if no test gas flows into the test volume through a leak, i.e. if the gas conditions at the gas detector do not change. The predicted measurement signal Í(t+t0) can then be used for comparison with the actual measurement signal I(t+t0) at time t0 in order to assess whether the same gas conditions prevail at the gas detector at time t0 as at time t. For this purpose, the difference between the predicted measurement signal Í(t+t0) and the actually measured measurement signal at time t+t0 is used to assess the leak. Conventionally, such an assessment is only possible after the system time constant has elapsed. The measurement signal predicted in this way is therefore also referred to in this description as the accelerated measurement signal.

[0015] The predicted measurement signal Í(t+t0) can be formed from the measured measurement signal I(t) by multiplying the measurement signal I(t) by a constant C2 with 0 < C2 < 1. The comparison of the real measurement signal I(t+t0) with the predicted (accelerated) measurement signal Í(t+t0) is performed by forming the difference between the measurement signal I(t+t0) and the accelerated measurement signal Í(t+t0).

[0016] According to the invention, the difference thus formed, e.g., I(t+t0) - C2 I(t), is used to assess whether the test object has a leak. For example, if the difference is particularly large, this can be considered an indication of a leak because the detector detects an additional, previously non-existent gas component. A particularly small difference can indicate unchanged conditions at the detector or a leak-tight test object. For this purpose, the difference can be compared, for example, with a threshold value. This enables accelerated leak testing or leak detection by knowing and utilizing the system time constant τ.

[0017] Preferably, the difference between the predicted measurement signal and the actually measured measurement signal at time t+t0 is multiplied by a constant C1 in order to numerically adjust the difference to the actual leak rate. C1 and / or C2 are real numbers, preferably C2 between 0 and 1 and C1 greater than 1. The measurement signal Í(t+t0) predicted for time t+t0 is determined by assuming an exponentially decaying curve for the measured signal I(t), which after the system time constant τ has elapsed at time t+τ is only approximately 36% of the value of the measured signal I(t). In other words, it is assumed that the signal I(t) decays exponentially and at time t+τ corresponds to a portion of the signal I(t) multiplied by a factor, where the factor includes the term 1 / e. For example, the signal I(t+τ) at time t+τ may only be 1 / e times the signal I(t) at time t.From such an exponentially decaying signal curve, for any time t+t0 with 0 <t0<τ der zugehörige Signalwert des Signals Í(t+t0) ermittelt werden.

[0018] For example, in the case of a measurement signal that increases in the presence of a leaking gas, such as in the case of a leaking gas partial pressure measurement (e.g., helium as a leaking gas), it can be assumed that a leak is present as soon as the difference is greater than a predefined threshold. The threshold can also be 0. In the case of a measurement signal that is reduced by the presence of a leaking gas, it can be assumed that a leak is present in the test object if the difference is less than a predefined threshold (e.g., 0).

[0019] Preferably, when comparing the difference with a threshold value, it can be checked whether the difference a) is above the threshold value, b) is in a range between zero and the threshold value, and / or c) is less than zero, i.e., negative. In case a), it can then be assumed that the test object has a leak. In case b), it can be assumed that the test object does not have a leak. In case c), it can be assumed that a fault is present, for example, in the form of an incorrect system time constant.

[0020] The threshold can be used to account for a background signal, for example, in the form of background noise in the measurement signal after the system time constant has elapsed. The background signal can result from signal noise and / or an offset signal. The offset signal can, for example, result from gas components that are emitted internally from the walls of the leak detection system. The threshold can be set, for example, as 5 to 10 times the value of the background signal or the background signal averaged over a period of time.

[0021] It is also advantageous if, when comparing the difference with a threshold value, a difference averaged over a period of time is used in order to ignore short-term outliers in the measurement signal.

[0022] The system time constant τ can be determined using a test leak, e.g. a spray leak, or it can be calculated from the volume of the test chamber and the pumping speed of the vacuum pump.

[0023] The system time constant τ is determined from the rate of rise or fall of the measurement signal, ie the measured leak rate signal, when a leak is switched on or off, ie when a leak is sprayed or the spraying of a leak is stopped.

[0024] The system time constant τ can be determined directly from the time it takes for the measured signal to decay to 1 / e of the signal generated or measured by a test leak, for example. The system time constant τ can be determined directly from the time it takes for the measured leak rate signal to rise to (1-1 / e) times the signal of a test leak.

[0025] Alternatively or additionally, the system time constant τ can be determined using the formula τ = t / Ln(I(t=0)) - In (I(t=t)), where I(t=0) is the measurement signal at the time of deactivation, removal or shutdown of a test leak and I(t=t) is the measurement signal at any time t after deactivation of the test leak. The system time constant can also be determined by using the formula given above as I(t=0) the measurement signal, which is the difference between the measurement signal when deactivating the test leak and before activating the test leak, and I(t=t) the difference between the measurement signal at time t after deactivation and before opening the test leak.

[0026] The basic idea is to technically accelerate the signal. The system's time constant is a fixed, technically measurable value and can be measured, for example, by switching a suitable test leak on and off from the leak detector itself. Due to the known time behavior, one can predict at any moment how the signal will behave when no test gas is being sprayed. This can be expressed mathematically as follows. I(t) :Signal leak rate at time t Í(t): Accelerated leak rate signal I(t - τ / n): Leak rate signal at time t - τ / n F: Leak rate suppression factor (0.9 to 0.999) n: Boost factor (how much higher the virtual pumping speed is compared to the physical one), n is a natural number τ: System time constant, which results, for example, from the effective pumping speed and volume. This must be measured by the system in advance.

[0027] The resulting signal Í (t) is compared to the leak rate signal I (t) accelerated by a factor of (n+1). In addition to the acceleration, a desired system time constant (the time at which the signal has risen to the final value due to spraying) can also be specified. n then results from the desired system time constant τ w as n = τ / τ w . The disadvantage of the filter is the increase in noise compared to the original signal. It is advantageous to adjust the spray duration of the test gas to the desired system time constant of the filter (or vice versa). In this case, the effective noise of the signal increases by only approximately 40%, regardless of the acceleration factor n. Furthermore, knowledge of the system time constant τ allows the filtering of the original leak rate signal Í (t) to be designed so that for larger system time constants I(t)The filter system time constant (the time range over which the average is taken) must also be adjusted. Since the system time constant represents the fastest signal response physically possible, averaging can be performed over a larger time range without slowing down the signal response. This can effectively reduce the filter noise. (Larger volumes require greater acceleration, but also allow a longer time span for signal averaging and thus lower noise.) 1) Determination of the system time constant for the leak rate measurement using a suitable method (internal test leak, external test leak, spray leak, input of volume and pumping speed) 2) If necessary, filtering of the leak rate signal adapted to the system time constant 3) Acceleration of the signal using the formula mentioned above 4) If necessary, adaptation of the desired system time constant (boost factor) to the spray duration of test gas or adaptation of the spray time to the set system time constant. Results:

[0028] A signal transformation of the form: I(t): Signal leak rate at time t I (t - τ / n):Leak rate signal at time t - τ / n F: Factor for the Leak rate suppression (0.9...0.999) n: Boost factor (how much higher the virtual pumping speed is compared to the physical one). τ: System time constant, which can result, for example, from the effective pumping speed and volume. This must be measured by the system in advance.

[0029] This leads to an acceleration of the measurement signal. The basic requirement is that the system time constant is determined beforehand, for example, using an internal test leak. Approach and basic idea

[0030] In vacuum leak detectors, the ratio of effective pumping speed to chamber volume determines the system time constant, which determines the maximum rate at which a signal can increase or decrease. Physically, a signal cannot increase or decrease faster, only slower due to, for example, permeation or outgassing from surfaces or pumping out of hidden volumes. If all slow effects are excluded, the signal behavior can always be described by a simple e-function.

[0031] For rising signals, the function looks like this: I t = I 0 1 − exp − t τ , for falling signals like this: I t = I 0 exp − t τ .

[0032] Since it is initially an exponential function, the maximum possible change in I(t) can always be determined using the signal I(t) (by taking its derivative). The maximum possible derivative (slope or decay) is always known and is derived from the leak rate signal.

[0033] The basic idea is to act as if the signal is currently falling at every point in the signal. This means that you always subtract the maximum physically possible change from the signal. Since this is an e-function, it makes sense to express it in units of the system time constant.

[0034] If you carry out the transformation you get the changed leak rate signal Í (t) .

[0035] In words, this means something like: Take the current signal and subtract the signal τ / n seconds ago, which can be expressed as a maximum decreasing change exp − t − τ n τ within τ / n. n is a natural number.

[0036] The equation can be simplified significantly due to its exponential nature.

[0037] Now you can check how the new signal behaves under different conditions. Rising signal:

[0038] I t = I 0 1 − exp − t τ I ˙ t = I 0 1 − exp − t τ − I 0 1 − exp − t − τ n τ e n

[0039] The ideal result is a time-independent signal with reduced amplitude. Time independence is always achieved when both the signal at the time t as well as the signal at the time t-τ should be on the rising edge of the signal. Before that, the signal rises exponentially. What does that mean? When helium is sprayed, the target amplitude is already reached after the specified time (τ / n). Although the signal is still rising, the filtered signal already outputs the final value. Typically, two system time constants are needed to almost reach the final value. A value of n = 1 halves the effective system time constant, so to speak; or to put it another way, it doubles the pumping speed. However, I would rather call this a virtual pumping speed. Values significantly larger than 1 can also be used. For example, a vacuum leak detector with the filter could easily achieve a virtual pumping speed of 200 l / s. The measurement signal becomes smaller and smaller as n increases.

[0040] So if you want to output the leak rate you have to measure the signal with 1 1 − 1 e n scale. The errors, or rather the noise, increase as a result. Falling signal:

[0041] I t = I 0 exp − t τ

[0042] By subtracting the expected signal, the result is exactly zero. However, the signal only becomes zero if both at the time t as well as at the time t - τ / n the signal is on the constant falling edge. When helium is sprayed, the signal remains almost exactly the period τ / n to see.

[0043] A value of zero is obviously not very useful for a leak detector. A small error in determining the system time constant τ can lead to negative leak rates and thus to the leaks being swallowed up. Ideally, the correction is scaled slightly with a parameter F to leave a small but visible leak rate signal.

[0044] An F of 0.9 reduces the indicated leak rate by one order of magnitude, a value of F of 0.99 by two. Stable signal:

[0045] I t = I 0

[0046] With a stable input signal, the filter always delivers a stable output signal. However, the signal level is reduced by a factor 1 − 1 e n lower. If you want to output the correct leak rate value, you have to scale the resulting signal accordingly. Filter summary

[0047] If you want to avoid negative leak rate values and also display the correct leak rate, the transformation of the signal into the filtered signal looks like this. I(t) :Signal leak rate at time t I (t - τ / n):Leak rate signal at time t -τ / n F:Leak rate suppression factor (0.9...0.999) n: Boost factor (how much higher the virtual pumping speed is compared to the physical one) τ: System time constant, which can result, for example, from the effective pumping speed and volume. This must be measured by the system in advance. Characteristics

[0048] The filter is designed to essentially accelerate the signal for spray leaks—that is, all signals whose changes result from the effective suction capacity and volume.

[0049] Permeation, hidden volumes, and background are not suppressed by the filter. On the one hand, this is a negative, as an offset is always visible; on the other hand, it makes it very easy for the user to distinguish between real leaks (capillary leaks) and permeation leaks. Especially for leak detectors with relatively low pumping speed (41) combined with large volumes, the system time constant for capillary leaks is quite large at 20 s. Since the filter maps the signal curve of capillary leaks, permeation leaks can be more easily distinguished, as they have a completely different (1 / root(t)) and slower signal curve. Basic requirement

[0050] The filter assumes that the system time constant is known with sufficient accuracy. In unknown systems, this can only be achieved by measurement. This is possible with a vacuum leak detector with an internal or external test leak. To do this, the test leak must be opened and you wait at least until the signal is reasonably stable. The test leak must of course be connected to the chamber. When the test leak is then switched off, the system time constant can be determined by measuring the decay time (e.g. to 1 / eth of the start signal). This can of course take up to a minute for large volumes and small effective pumping speeds, but it also provides the user with information about the system time constant and thus about the required spraying time at the leak. The time invested in determining the system time constant is already recouped within a few measuring points due to the significantly shorter measurement time. Rush

[0051] The filtered signal will contain significantly more noise than the original leak detector signal. This is partly because at least two signal values are required at any given time, and partly because the useful signal becomes smaller as n increases. However, it should be noted that very small leak rate indicators are necessary for leak detectors, as the signals are low at high volumes and short spray times. However, this is compensated for by the filter. Thus, the higher background noise may well be less disruptive.

[0052] The following figures illustrate exemplary embodiments in more detail. They show: Fig. 1 is a diagram showing the resulting signal curves of a first embodiment, Fig. 2 is a schematic representation of the vacuum leak detector, Fig. 3 is a diagram for determining the system time constant of the system, Fig. 4 is a further diagram for determining the system time constant, Fig. 5 is a further diagram for determining the system time constant and Fig. 6 is an example of filtering to illustrate the accelerated signal and the system time constant.

[0053] In the embodiment according to Fig. 1 The upper curve shows the measurement signal of the gas detector and the lower curve the accelerated signal Í(t) for a volume of the test chamber and the gas-conducting connection between the test chamber and the gas detector of 100 l, a suction speed of the vacuum pump of 4 l per second and an acceleration factor (boost factor) n = 1. The comparison of the lower curve with the upper curve according to Fig. 1shows that the accelerated signal curve Í(t) increases significantly faster than the actually measured curve I(t).

[0054] Fig. 2 shows a schematic representation of the vacuum leak detector, which consists of a test chamber 12, a gas detector 14, and a vacuum pump 16. The gas detector 14 is connected to the test chamber 12 via a gas line 18. The test chamber contains a test specimen 20 filled with a test gas. The gas detector 14 is electronically connected to an evaluation unit 22, which receives and processes the electronic measurement signal generated by the gas detector.

[0055] In another example, a UL1000 was used with a 50-liter drum. A limp valve combined with a TI4-6 was used as the adjustable test leak. The data input for the filter was the leak rate signal combined with the fixed filter, both to avoid the influence of different filter times and to measure noise amplification.

[0056] Various tests were conducted by opening and closing the limp valve, and the signal was examined for different acceleration levels. The signal drop and increase were particularly examined, as was the question of whether the leak rate predicted by the filter was accurate or to what extent it deviated. The filter was tested in the range between 1x10 -3 < and 1x10 -9 < mbar l / s. 1) The filter is working. 2) The signal can technically be accelerated by a factor of 32, allowing a virtual suction capacity of commercially available leak detectors of almost 1000 l / s. 3) The system time constant required for the filter can be determined using the internal test leak and will take between 15 and 20 s. 4) The accelerated signal has noise that is 1.4 times the acceleration factor higher than the input signal, which is in line with the theoretical assumption. 5) For short spray bursts (short compared to the system time constant), the noise only increases by a factor of 1.4 as long as the acceleration factor is not too high and is appropriate for the spray time. 8) The leak rate prediction is good and the filter over- / underdrives the signal very little.9) Smoothing the input signal to match the actual system time constant significantly improves the noise without significantly degrading the system time constant. 10) The waiting time for spraying after a large leak can be reduced by up to one minute for a 50-liter volume.

[0057] Fig. 3 illustrates an example for determining the system time constant of the leak detection system by measuring the slope or slope of the logarithmic measurement signal.

[0058] Fig. 4 shows an example of a direct determination of the system time constant by measuring the time that elapses until the measurement signal has dropped to a 1 / e-th fraction.

[0059] Fig. 5 illustrates an example for the direct determination of the system time constant by measuring the time until the measurement signal has risen to a (1-1 / e)th fraction.

[0060] Fig. 6shows an example of filtering the measurement signal. A leak is sprayed with 10 -7< mbar l / s. The measurement signal I(t) is shown in blue / dashed lines. The accelerated, transformed signal Í(t) is shown in red. The system time constant of the dashed original signal I of t is 17.5 seconds. The system time constant of the transformed signal Í(t) is in Fig. 6 from left to right for acceleration factors n=1.75; 3.5; 5.83; 8.75; 17.5; 35: 10; 5; 3; 2; 1; 0.5 seconds.

Claims

1. A method for evaluating the measuring signal of a vacuum leak detector comprising a vacuum pump (16) and a test chamber (12) connected with the vacuum pump (16) and a gas detector (14) connected with the test chamber (12), characterized by determining the system time constant τ of the vacuum leak detector, wherein the system time constant τ is the duration of the system-related rise of the measured signal in response to a gas component escaped through a leak in a test object or the duration of the system-related drop of the measured signal in response to a gas component no longer escaping from a leak in the test object, followed by the steps of: generating a measuring signal I(t) of the gas drawn from the test chamber (12) containing the test object at a time t, using the gas detector (14), buffering the measured signal I(t), forming a measuring signal Í(t+t0) prognosed for a future time t+t0, based on the measured value of the buffered measuring signal I(t) and the system time constant τ, where t0 < τ, generating a measuring signal I(t+t0) of the gas drawn from the test chamber (12) at a time t+t0, using the gas detector (14), forming the difference between the measuring signal I(t+t0) and the prognosed measuring signal Í(t+t0), judging, whether the test object has a leakage, based on the difference formed.

2. The method according to claim 1, characterized in that a leak in the test object is considered as detected, if the difference is greater than a threshold value.

3. The method according to claim 1 or 2, characterized in that the test object is considered to be tight, when the difference is smaller than or equal to a threshold value and greater than or equal to zero.

4. The method according to one of the preceding claims, characterized in that an error is assumed, for example in the form of an inaccurate system time constant, when the difference is smaller than zero.

5. The method according to one of claims 2-4, characterized in that the threshold value is assumed to be r-times the value of a background signal of the measuring signal, where r is a rational number greater than zero and preferably greater than or equal to 5 and / or smaller than or equal to 10.

6. The method according to one of the preceding claims, characterized in that the accelerated measuring signal Í(t) is formed from the measuring signal I(t) by the following transformation: using a previous measuring signal I(t') of a previous time t'=t-τ / n, where n is a rational number greater than 0, multiplying the previous measuring signal I(t') by a second constant C2, forming the difference I(t) - C2 I(t') between the measuring signal I(t) and the previous measuring signal I(t') scaled by the second constant C2, multiplying the difference I(t) - C2 I(t') by a first constant C1, to thereby obtain a measuring signal Í accelerated by the factor n, which corresponds to a measuring signal compensated by the system time constant τ.

7. The method according to claim 6, characterized in that the second constant C2 is a positive real number smaller than 1 and corresponding preferably to the reciprocal of the n-th root of Euler's number e or including this reciprocal.

8. The method according to claim 6 or 7, characterized in that the first constant C1 is a real number greater than 1 and preferably includes or corresponds to the term 1 / 1 − 1 / e n .

9. The method according to one of claims 6-8, characterized in that the second constant C2 includes the n-th root of Euler's number e and preferably includes the reciprocal of the n-th root of e.

10. The method according to one of claims 6 - 9, characterized in that the second constant C2 is multiplied by a factor F, where F is a rational number <1 and is preferably between 0.9 and 0.999.

11. The method according to one of the preceding claims, characterized in that the accelerated signal Í(t) is calculated using the formula: where I(t): signal of leak rate at time t, Í(t): accelerated leak rate signal, I(t - τ / n): signal of the leak rate at time t - τ / n, F: factor for leak rate suppression, n: boost factor, n is a natural number, τ: system time constant.

12. The method according to one of the preceding claims, characterized in that the system time constant τ is determined by means of a test leak which may be an internal test leak of the vacuum leak detector or an external test leak that can be connected with the test chamber (12).

13. The method according to one of the preceding claims, characterized in that the vacuum time constant of the vacuum leak detector is calculated as the system time constant τ from the volume of the test chamber (12) and the pipeline connecting the test chamber (12) with the gas detector (14) and from the suction capacity of the vacuum pump (16).

14. The method according to one of the preceding claims, characterized in that the system time constant τ is determined from that time period which elapses from the time of switching off, deactivating or removing a test leak from the vacuum leak detector to the decay of the measuring signal to a predetermined value, or which elapses from the time of switching on, activating or adding a test leak to the vacuum leak detector to the rising of the measuring signal to a predetermined value.

15. The method according to one of the preceding claims, characterized in that the predetermined time corresponds to the time at which the measuring signal corresponds to 1 / e times the steady measuring signal of the test leak.

16. A vacuum leak detector for performing the method according to one of the preceding claims, comprising a vacuum pump (16), a test chamber (12) connected with the vacuum pump (16) and a gas detector (14) connected with the test chamber (12), characterized by an evaluation unit (22) for evaluating the measuring signal of the gas detector (14), wherein the evaluation unit (22) is configured to perform the method according to one of the preceding claims.

17. The vacuum leak detector according to the preceding claim, characterized in that the evaluation unit (22) comprises a memory in which the process steps are stored.

18. The vacuum leak detector according to one of claims 16 or 17, characterized in that the evaluation unit (22) comprises a microcontroller configured to perform the method in an automated manner.