British Journal of Radiology (2004) 77, 745-750
© 2004 British Institute of Radiology
doi: 10.1259/bjr/92529963
Considerations on the measurement of practical peak voltage in diagnostic radiology
F J Ramírez-Jiménez, ENG
R López-Callejas, PhD
J S Benítez-Read, PhD
and
J O Pacheco-Sotelo, PhD
Instituto Nacional de Investigaciones Nucleares, Direcciones de Servicios Tecnológicos e Investigación Científica, Apartado Postal 18-1027, Colonia Escandón, C.P. 11801, México D.F., México and Instituto Tecnológico de Toluca, Departamento de Electrónica, Apartado Postal 890, Toluca, México
 |
Abstract
|
|---|
An evaluation of the non-invasive measurement of the practical peak voltage (PPV) in the quality control of X-ray units used in diagnostic radiology was carried out. Two instruments were employed: the PTW Diavolt Universal Tester with readings in PPV and the Waveform Tester for X-rays (WATEX) prototype proposed here, which uses a PIN structure (P-type diffusion, Intrinsic region, N-type diffusion) photodiode as a sensor. The reference for the measurements was the voltage signal obtained in an oscilloscope from an invasive high voltage divider in order to verify the accuracy and precision of the measurements. The readings of the PPV in the Diavolt show a systematic error between 1% and 8%, always being less than the real value. An explanation for this difference is proposed, based on the relation between the effect of the X-rays in the film and the response of the sensor to the product of the applied voltage to the X-ray tube (peak voltage kVp), and the anode current. This explanation was confirmed using the WATEX waveform tester.
|
Introduction
|
|---|
Digital radiography is becoming a standard technique, due to better quality of images and reduced patient dose. In order to avoid unnecessary radiation dose to the patient, recent experience [1] has shown that there is a need to consider many factors, some of which were not considered in analogue techniques. It is important to keep the radiation doses to patients at a minimum, because the use of X-rays in medicine for diagnosis represents the largest man-made source of public exposure to ionizing radiation [2].
Quality control of X-ray units is necessary to ensure good performance and minimum dose to patients. In quality control, several parameters are measured. According to several recommendations [3, 4], the main parameters are: (a) high voltage applied to the X-ray tube; (b) exposure time; (c) efficiency of the X-ray unit, calculated as the ratio of the absorbed dose to the product of current and time [Gy/mA-s]; (d) the quality of the beam, measured as the half value layer (HVL) using aluminium filters; (e) the air kerma rate (Gy min–1); and (f) the absorbed dose (Gy).
One of the most important parameters is the tube voltage, because this parameter strongly influences the dose to patients and the exposure of the film [5].
In Secondary Standard Dosimetry Laboratories, several instruments are used as secondary standards for the precise measurement of dose and air kerma due to X-rays [6]. The certified instruments for dosimetric measurements are ionization chambers for the different ranges of energy, electrometers for the measurement of the ionization chamber currents and several non-invasive X-ray meters. To guarantee the required accuracy of results, several parameters are important such as voltage waveform, tube voltage, anode tube current, anode angle and inherent and added filtration.
An evaluation of the non-invasive measurement of the practical peak voltage (PPV) in the quality control of X-ray units used in diagnostic radiology is the subject of our work. Two instruments were employed as X-ray testers: (a) the Diavolt Universal Tester (PTW, Freiburg, Germany) [7] whose calibration is based on the definition of PPV [8]; and (b) the Waveform Tester for X-rays (WATEX), which is a prototype we propose that uses a PIN diode as a sensor. An invasive instrument, the Dynalyzer High Voltage Divider (Radcal Corp., Monrovia, CA) [9], was used in this study as a reference to make measurements of tube voltage and anode current of the X-ray unit. Using the precise waveform obtained with the Dynalyzer, an evaluation was made of the concept of PPV from a new point of view.
|
Basic considerations
|
|---|
The effect of X-rays on the film depends upon three factors: PPV, the intensity of X-rays and the contrast equivalent X-ray tube voltage. Each of these will be considered as follows.
Practical peak voltage (PPV)
In diagnostic radiology, the measurement of the X-ray tube voltage has always been problematical because the term "peak voltage" is not well defined. This could indicate the maximum voltage or the arithmetic mean. In order to define a precise quantity for comparisons of voltage related to the final effect on the film, the PPV concept has been proposed by Kramer and collaborators [8].
The PPV is defined as the equivalent value of a voltage of any waveform related to an ideal X-ray generator that provides a constant voltage and that produces the same contrast on the film. With this definition, the waveform of the X-ray generator is not important for the evaluation of the results on the film and any comparison between different X-ray machines could be carried out in these terms.
The PPV (Û) can be calculated from the instantaneous values of the tube voltage by using the following equation [8]: 
where p(Ui) is the probability of occurrence of any voltage Ui in the range from [Ui–(
U/2)] to [Ui+(U/2)]; and w(Ui) is a weighting function obtained theoretically in [8]. When the measurement is carried out at a constant rate and with a fixed conversion time, the probabilities p(Ui) are the same for all the Ui values, therefore Equation (1)
can be simplified to:
Kramer and collaborators [8] have demonstrated that Û does not depend strongly on the X-ray tube characteristics, such as anode angle and filtration used, nor on the contrast configuration employed; therefore, this concept can be generalized to nearly all the normal conditions encountered in diagnostic radiology applications.
Intensity of the X-rays
The contrast obtained in the film is expected to be related to the intensity of the X-rays. The intensity of the X-rays In is defined as [10]:
where N is the number of photons per second, E is the energy of photons in keV, A is the area in cm2, and 
is the X-ray fluence in number of photons per second per unit area. The intensity of X-rays generated is related to the operational parameters established in the X-ray unit by [10]:
where k is a constant, i is the current through the tube in amperes, and V is the accelerating voltage applied to the tube in volts.
Contrast equivalent X-ray tube voltage
The contrast, CK, is defined [8] as the ratio of the optical density in two different areas of a film under different conditions. The contrast depends on the tube characteristics, the contrast geometry and the total filtration. In terms of measurements, the contrast is calculated as the ratio of the air kerma measured behind a blank phantom and the air kerma behind a phantom covered with a contrast material:
The air kerma K is given as [11]:
where
is the mass energy transfer coefficient for air. The effect on the film is proportional to the deposited energy; then, the contrast can be calculated as:
where Ei is the i-th energy of the photons, n is the number of energies considered, µp(Ei) and µc(Ei) are the linear attenuation coefficients of the phantom and the contrast element, respectively, and dp and dc are the thicknesses of the phantom media and the contrast element, respectively. In the definition of PPV, the equivalence of different tube voltages to the X-ray tube was obtained by reference to the same contrast produced on the film. Baorong et al [12] have verified experimentally that an aluminium plate of 1 mm thickness can be used as a contrast element without any loss of generality.
|
Experimental set-up
|
|---|
The waveform of the X-ray exposure and exposure time were measured with the arrangement shown in Figure 1. The X-ray unit used was an MPH 65 (General Electric, Waukesha, WI) with a tungsten target, an intrinsic filtration of 1.5 mm of aluminium and a high voltage generator of the high frequency type. The Dynalyzer IIA is a precision instrument that was used as a reference to make measurements of the applied voltage and anode current in the X-ray unit. The Dynalyzer is insulated in sulphur hexafluoride gas with buffer amplifying circuits to provide isolated analogue signals proportional to the anode voltage, cathode voltage, anode plus cathode voltage, anode current and filament current. The Dynalyzer is factory calibrated to operate in the range from 20 kV to 150 kV; it is an invasive instrument and must be connected in series between the high voltage generator and the X-ray tube. It provides analogue outputs with a ratio of 1 mV mA–1 for the anode current and 0.5 V/10 kV for the anode plus cathode voltage.
Two instruments were employed as X-ray testers in the configuration illustrated in Figure 1. One was the Diavolt Universal Tester, and the other the Waveform Tester for X-rays (WATEX), which is our proposed prototype. The X-ray testers were positioned at a reference distance of 50 cm from the focus of the X-ray tube to ensure sufficient X-ray intensity.
The Diavolt can measure the exposure time and the voltage applied to an X-ray unit in the range from 22 kV to 150 kV. The measurement of the tube voltage is based on the principle that the ratio between the responses to radiation of two detectors with different filtration is directly proportional to the voltage applied to the X-ray tube [13]. A heavy filtration of several millimetres of aluminium, molybdenum or other materials is used in both detectors inside the instrument to reduce the influence of different inherent filtration encountered in the field with practical X-ray units. The generated signals are processed by an intelligent circuit. The measurement can be programmed to be in kVp, average voltage, or PPV values.
The basic WATEX has a response directly proportional to the energy delivered by the X-ray unit. The operational principle of the WATEX is based on the response of a PIN diode to the X-rays. The PIN diode is connected in the photovoltaic mode (see Figure 2).
We used an OPF420 PIN diode as the X-ray detector in the WATEX prototype. It was tested for this application in the range from 25 keV to 120 keV; it was selected with a low reverse current in the dark condition. The first stage of the circuit converts the current generated in the detector to voltage with a conversion gain of 20 mV nA–1, thus R1=20 M
. In the second stage of amplification, a voltage gain of 12 is realised with R2=1 k and R3=12 k. With the obtained global gain, a response to all the different combinations of setting of current and voltage in the tube was assured. The operational amplifier used (LF441) was selected because of its low input current, low offset voltage at the input and low power consumption.
The signals from the Dynalyzer and the X-ray testers were measured with a digital oscilloscope that was used to save the image of the X-ray exposure waveform and also as a precision digitizer with an analogue to digital converter (ADC) of 8 bits. The data were also transferred to the personal computer (PC) for further analysis. We used Tektronix TDS220/TDS240 oscilloscopes (Tektronix, Beaverton, OR) with an accuracy of ±2% and a maximum sampling rate of 1 G samples s–1. The number of samples depends on the setting of the time base employed [14].
|
Results
|
|---|
The accuracy and precision of the settings of the General Electric MPH 65 X-ray unit were evaluated, taking as reference the calibrated Dynalyzer voltage divider. The results of the measurements of tube voltage and time accuracy in the range from 45 kV to 110 kV are shown in Figure 3. We can see that the voltage error is less than 6% and the time error is less than 0.45%. The measured repeatability of the X-ray unit was also good.
View larger version (16K):
[in this window]
[in a new window]
|
Figure 3. Time and voltage error of the X-ray unit in the settings, for 100 mA and 250 ms, corresponding to 25 mAs.
|
|
The Diavolt has three measuring ranges for conventional radiology applications, 40–85 kV, 65–110 kV and 80–150 kV. Time accuracy of the Diavolt was calculated from the analogue signal obtained by the oscilloscope and from the digital reading. The results can be seen in Figure 4, for the three ranges. Time measured from the analogue output has an error of approximately 4.5% and from the digital reading has an error less than 1.5%.
View larger version (20K):
[in this window]
[in a new window]
|
Figure 4. Time error of the Diavolt for an X-ray exposure of 100 mA and 250 ms. Upper traces are for the time measured in the analogue output signal, lower traces are for the digital readings.
|
|
A measurement of the error in PPV, peak kilovoltage (kVp) and mean value readings was made. The results are shown in Figure 5a for PPV and in Figure 5b for kVp. The mean value readings have the same error as for peak kilovoltage. All the voltage measurements in the Diavolt show a systematic error between 1% and 8%, always being less than the true value. From Figure 5 it is concluded that the values of PPV, peak kilovoltage and mean voltage, measured in the Diavolt, are almost linearly related.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 5. Error in the measurement of tube voltage in the Diavolt for 100 mA, 250 ms: (a) practical peak voltage (PPV); (b) kVp.
|
|
It was also observed that the Diavolt showed some "black holes" in which no readings were obtained; this could reduce the degree of confidence of the instrument.
The error in the measurement of exposure time of the X-ray unit with the WATEX and the oscilloscope are shown in Figure 6. The error obtained was less than 0.45%.
View larger version (16K):
[in this window]
[in a new window]
|
Figure 6. Time error in the measurements of exposure time with the WATEX for a shot of X-rays, 100 mA and 250 ms.
|
|
|
Discussion
|
|---|
The difference encountered in the readings of the Diavolt with respect to the Dynalyzer led us to look for an explanation. We decided to compare the waveforms obtained from the Dynalyzer, the Diavolt and the WATEX tester. These can be seen in Figure 7. It can be observed from Figure 7, trace 1, that the voltage waveform of the X-ray unit is relatively constant along the exposure time axis. Therefore the X-ray unit used can be considered almost as an ideal constant voltage unit.
View larger version (34K):
[in this window]
[in a new window]
|
Figure 7. Response of different instruments to a 80 kV, 100 mA X-rays exposure: trace 1 is the signal of anode to cathode voltage measured with Dynalyzer; trace 2 is the anode current signal measured with the Dynalyzer; trace 3 is the output signal from the Diavolt; trace 4 is the output signal from the WATEX.
|
|
The oscilloscope output of the Diavolt, Figure 7, trace 3, has a timing delay that is not related to the delay established in the programming of the instrument. The results obtained for the voltage error in Figure 5 were consistent for long exposure times, indicating that this error is not dependent on the delay observed. The oscilloscope output of the Diavolt is a processed signal that is directly proportional to the PPV measured as defined in Equation (2); this proportionality was confirmed by measurements (see Figure 8).
View larger version (19K):
[in this window]
[in a new window]
|
Figure 8. Relationship between the output voltage of the Diavolt and the tube voltage, measured for 100 mA and 250 ms.
|
|
The signal from the WATEX tester (Figure 7, trace 4) is the straight response of the sensor and shows a finite response time and a settling time similar to the one obtained in the waveform of the tube current (Figure 7, trace 2). It is clear that the oscilloscope output of the Diavolt does not have a response similar to the response of the WATEX tester, due to the fact that the output of the Diavolt is a processed signal.
In real applications, the relatively slow response of the current in the high voltage generator includes a reducing factor in the calculation of the PPV, whereas the finite response time of the sensor includes an error in the calculations of the PPV as measured in the Diavolt. The responses of the film and of the detector are proportional to the X-ray intensity as defined in Equation (4) in terms of voltage and current in the X-ray tube and not only due to the tube voltage used. The basic detector response is proportional to the intensity of the X-rays as defined in Equation (3) and to the kerma as in Equation (6). To verify this statement, related to the detector itself, a comparison was performed between the normalized responses of instruments that measure exposure rate and the WATEX tester. The instruments considered were: a calibrated PTW 365-ionization chamber, a Rad Check Plus model 06-526 meter manufactured by Victoreen (Cleveland, OH), and a 10X5-60 ionization chamber by Radcal (see Figure 9). All the instruments present the same type of response.
View larger version (21K):
[in this window]
[in a new window]
|
Figure 9. Comparison of the normalized response of four instruments following an X-ray exposure, 100 mA.
|
|
|
Conclusions
|
|---|
The characteristic shape of the current in the X-ray unit, as indicated by the Dynalyzer trace in Figure 7, affects measurements made by the Diavolt for the PPV according to Equation (2). This is one of the reasons for the consistently low readings from the Diavolt. Furthermore, considering that the sensor in the Diavolt has a finite response time as indicated in Figure 7, trace 4, this introduces an additional error in the measurement of the PPV. The readings obtained for PPV by the Diavolt therefore demonstrate a systematic error of between 1% and 8%.
The evaluation of the Diavolt showed that the device is accurate enough to be considered as a field reference in measurements for quality control in diagnostic radiology. The PPV reading is quite useful when we want to compare different operating parameters that can produce the same quality in the film image in terms of the contrast as defined in Equation (7).
The WATEX tester proposed was useful for verifying the waveform obtained in a basic detector system, showing a response proportional to the X-ray intensity and to the kerma. This relationship helped to explain the apparent error in the measurements realised with a device calibrated in terms of PPV.
|
Acknowledgments
|
|---|
The help of the Radiation Metrology Laboratory of ININ for the use of their X-ray units and the support from the IAEA in the ARCAL LIII project (Regional Arrangements of Cooperation for the Promotion of the Nuclear Science and Technology in Latin America and the Caribbean) are acknowledged.
Received for publication August 26, 2003.
Revision received February 12, 2004.
Accepted for publication April 6, 2004.
|
References
|
|---|
- Freudenberger J, Hell E, Knupfer W. Perspectives of medical X-ray imaging. Nucl Inst and Meth in Phys Res. Section A. 2001;466:99–104.
- Fung KKL, Gilboy WB. The effect of beam tube potential variation on gonad dose to patients during chest radiography investigated using high sensitivity LiF:Mg,Cu,P thermoluminescent dosemeters. Br J Radiol 2001;74:358–67.[Abstract/Free Full Text]
- National Council on Radiation Protection and Measurements. Quality assurance for diagnostic Imaging Equipment. Bethesda, USA. NCRP Report 99, 1988.
- Minister of National Health and Welfare. Diagnostic X-Ray Equipment Compliance and Facility Survey. Canada, 1994.
- Bushong SC. Radiologic Science for Technologists (5th edn). London: Mosby-Year Book, Inc., 1993:156–70.
- Mann WB, Rytz A, Spernol A, McLaughlin WL. Radioactivity measurements, principles and practice. London: Pergamon Press, 1988:26–8.
- PTW Freiburg. Diavolt Universal Tester, User Manual, PTW Freiburg, Germany, 2001.
- Kramer HM, Selbach HJ, Iles WJ. The practical peak voltage of diagnostic X-ray generators. Br J Radiol 1998;71:200–9.[Abstract]
- Radcal Corporation. Dynalyzer II A High Voltage Unit. Technical Manual, USA. 1989.
- Tertian R, Claisse F. Principles of quantitative X-ray fluorescence analysis. London: Heyden & Sons, 1982:36–9.
- http://physics.nist.gov/PhysRefData/XrayMassCoef/, X-Ray Mass Attenuation Coefficients.
- Baorong Y, Kramer HM, Selbach HJ, Lange B. Experimental determination of practical peak voltage. Br J Radiol 2000;73:641–9.[Abstract]
- Gambaccini M, Rimondi O, Marziani M. An electronic instrument for high-voltage measurements of an X-ray unit for mammography. Report 18. London: British Institute of Radiology, 1998:76–9.
- Tektronix. TDS 220/240 series Oscilloscopes. User Manual.
This article has been cited by other articles:
|
|
|
|
 
H-M KRAMER and H-J SELBACH
Extension of the range of definition of the practical peak voltage up to 300 kV
Br. J. Radiol.,
September 1, 2008;
81(969):
693 - 698.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
