Challenges to the validity of radiometric dating are based on ignorance and misinformation
One of the main objections to radiometric dating is that radiometric ages do not agree with each other or that contamination renders ages meaningless or that the half-life of a chemical element is variable.
With respect to inconsistent readings, the claim is partially true. Early mass spectrometers were not as sensitive as machines today and the methods for separating, cleaning and analysis were less sophisticated. Although ye-creationists like Snelling talk about contamination of isotopic systems as if it were a foreign concept to modern geology, most geochronologists routinely check for possible contamination using a variety of methods. In addition, geologists recognized that rocks could be contaminated with excess daughter or parent or loss of parent/daughter that would also affect the age as determined by radiometric methods.
Creationists have seized upon these discoveries and held them forth as evidence that radiometric dating is inaccurate. But is this the case? Simply put each radiometric system is based on the assumption that each system has a different half-life (derived from the decay 'constant' which is simply the length of time it takes for 1/2 of the radioactive parent to decay to a stable daughter). In addition to variable half-lives, each mineral will 'close' at different temperatures (closure, is simply defined as the point where no daughter/parent is lost or gained*).
There are a number of different methods that geologists use to check for loss/gain and these are incorporated into most analyses (isochron methods, stepwise degassing etc). If radiometric decay rates are not constant and rocks behave as open systems, it would be the exception, rather than the rule, for ages to agree with one another.
Here are some http://gondwanaresearch.com/radiomet.htm examples of radiometric age determinations on the same rocks (using different isotopic methods).