Para Uma Mesma Intensidade De Radiação

How to Maintain Consistent Radiographic Exposure When Thickness Changes

When you're doing industrial radiography, the most common headache is keeping the same film density or detector signal when you move from a thin section to a thick one. The inverse square law and attenuation through the test piece mean you can't just set it and forget it. You have to recalculate. The core principle is straightforward but easy to get wrong in practice. If you need the same radiation intensity reaching the detector, you're essentially solving for exposure compensation across a change in material thickness or source-to-object distance. The basic equation involves the exposure factor, which is a product of milliampere-seconds and the square of the source-to-film distance, divided by a reference value. In formula form: EF = (mA × time) / SD². When you change thickness, you adjust mA or time to keep the exposure factor at the level your film or detector expects. I used to work on a pipeline welding inspection job where we had a mix of 8mm and 22mm thick sections in the same run. The procedure called for a single exposure setting, which was impossible. I ended up calculating two separate exposure factors using the gradient or k-value method from the chart. For the thinner sections, I dropped the time from 40 seconds down to about 12 seconds at 250 kV, and for the thicker joints I pushed up to roughly 90 seconds at the same voltage. The density across both came out within ±0.15 of the target, which is inside the acceptable range per the code. Skipping that calculation would have left the thin joints overexposed and the thick ones underexposed, and the welder would have had all the reasons to complain.

Here's what most people miss: the gradient method assumes a straight-line relationship on a semi-log plot between thickness and exposure factor. That holds pretty well for steel in the 100–400 kV range, but it breaks down if you're dealing with high-energy gamma sources or very thick sections where the spectrum hardens significantly. When I ran into that on a 60mm steel plate inspection with an Ir-192 source, the calculated exposure was off by nearly 30% because the beam quality shifted as it passed through the material. The workaround was to do a test exposure on a step-wedge block cut from the same material batch, measure the actual density, and then back-calculate an empirical correction factor. It added about 45 minutes to the setup but saved us from re-radiographing three entire pipe spools. Another thing nobody tells you about distance compensation: the inverse square law only applies to the source-to-detector path. If you're changing the source-to-object distance to avoid hot spots or improve geometry, you also need to account for the change in effective source-to-film distance. I've seen technicians calculate the mA adjustment using the source-to-object distance by mistake, which threw the exposure off by a noticeable margin. Always use the source-to-detector distance in the denominator, not the source-to-part distance. For digital detectors, the concept is similar but the implementation differs. Digital systems have a wider dynamic range than film, so small miscalculations don't ruin the image the way they would on film. However, the system's automatic exposure control can mask problems. I've seen cases where the AEC terminated the exposure based on a reading from a small area of the detector that happened to see less attenuation, leaving the rest of the image underexposed. The fix is to override the AEC and manually set the exposure based on your calculated factor, then verify with a phantom or IQI in the field of view.

If you're working with a material other than steel, the attenuation coefficients change. Aluminum, titanium, and copper all require different exposure factors at the same thickness. The k-values from your exposure chart are material-specific, so swapping materials without adjusting the chart values will give you incorrect results. I once had a project with titanium pressure vessels where someone used the steel k-value table. The first radiograph showed adequate density on the IQI but the weld profile was completely washed out because the actual attenuation in titanium was significantly lower than in steel at the same nominal thickness. The practical takeaway is that you need a reliable exposure chart for your specific source energy, material, and film or detector type, a method to interpolate or extrapolate between chart values, and a verification step using a test coupon whenever you leave the standard conditions. Without the verification step, you're guessing, and in radiography guessing costs money and time. For a quick reference, here's a simplified exposure factor calculation workflow:

Step one: determine the thickness of the section you're radiographing and look up the corresponding exposure factor on your chart. Step two: if you're changing the source-to-film distance, recalculate using the inverse square relationship. Step three: solve for the new milliampere-seconds value. Step four: run a test exposure on a representative sample and measure the resulting density or detector signal. Step five: adjust and re-test if the result falls outside acceptable limits. This process typically takes between 20 and 40 minutes for the first setup on a new material or thickness range, after which subsequent exposures in the same family come much faster because you've already established the correction curve. On a typical day with multiple thickness variations, that upfront investment saves roughly two to three hours of rework compared to the trial-and-error approach most crews fall back on when they don't bother with the calculation. One final note: if your procedure requires strict adherence to codes like ASME Section V or ISO 17636, document every exposure factor calculation and the corresponding verification result. Auditors will ask for it, and having the paperwork ready cuts review time from hours to minutes during the final report stage.