Testing Nonparametric Tukey HSD

Hello,

I have written a little script to calculate a nonparametric Tukey HSD following Kruskal Wallis.  I need something to test it against to see if it performing correctly.  It does ok vs. hand calculations on a small data set, but I would like something more robust to test with.  Does anyone have something to use as a benchmark?  That would mean either a data set with known results, or another Tukey calculator.

Thank you

Tom Georgianna