An opinion by Dr. Jon Shapiro, President, Greenwich Instrument Co. Inc.
There are many reasons why a properly calibrated Dynalyzer or Voltage Divider will read differently from a non-invasive KVP meter. The differences are even larger in Mammo voltage ranges, where a 1 KVP difference in readings in not only a 4% error, but makes large differences in image quality.
A voltage divider is not filtration dependent. If you doubt that filtration makes any difference between invasive and non-invasive devices, put a thin piece of aluminum foil in the beam over a non-invasive device, and see how the readings change. A 10 micron change in the MO (molly) filtration will change the non-invasive readings by more than 1 KVP at 27 KVP!!
Most non-invasive KVP meters have problems if there is significant high frequency ripple. If the waveform is smooth, then this should not be a problem. If the instrument has a three phase, single phase switch, it means it has problems with waveforms and ripple. It has to add a fudge factor.
Voltage dividers can be a source of error from lack of understanding in hook up, such as impedance mismatch, or limited frequency response. Please read Application Note 6 for details.
The Dynalyzer has few limits to its accuracy, other than miscalibration, or failure. The scope outputs have a flat frequency response to 100 kHz, at which point the internal amplifiers roll off. The digital display has good frequency response to 10 kHz, beyond which there is some error in reading the high frequency ripple. Usually the ripple is a few percent, and the error is a percentage of this, so in practical uses, it is not a problem. In any case, at a frequency that the Dynalyzer would have a minor problem with, the non-invasive device detectors and amplifiers would be brain dead, in my opinion.
While a non-invasive meter is labeled KVP, it is calibrated against a calibrated divider or Dynalyzer by the manufacturer, and it is calibrated with a specific x-ray tube ( beam quality) on a specific x-ray generator (waveform). Without a doubt, a brand new instrument would read differently on different generators the very day it was made.
GiCi believes that non-invasive meters have a place in the quality control market. Their best use is as a double check against the voltage dividers. While setting up the x-ray generator, use the non-invasive to make some comparison readings while the Dynalyzer, Inspec 200 or voltage divider is in the circuit. Then remove the Dynalyzer. If the KVP reading on the non-invasive changes, there is the possibility of "loading" due to the presence of the Dynalyzer. Then follow this procedure.
Set up generator with Dynalyzer in Anode and Cathode.
Remove Cathode from Dynalyzer and connect directly to tube. Leave the anode in.
Re check and readjust mA stations. Take some non-invasive vs Dynalyzer or divider comparison readings and write them down.
Remove Dynalyzer or divider from circuit. See if the comparison readings from step 3 have changed significantly. Note with high frequency generators, please use HV test cable of 5 feet or less. High frequency generators are affected to some extent by the extra capacitance of loading due to long cables.... some more than others! If the readings are consistent, write them down. The Inspec 100 and 200 x-ray high voltage calibration systems have been designed to minimize the loading effects due to the extra high voltage cables required with Dynalyzers or voltage dividers. These devices plug right into the generator or tube HV sockets, and the existing HV cable is connected to the Inspec. Write, call or e-mail for information.
Use your comparison readings to allow period QC non-invasive tests on your generator. Even if the non-invasive readings differed from the generator control settings, they are a reference, and should they change in the future, you can then take corrective action.