Friedman post hoc Conover - 2 formulas?

#1
I'm trying to figure out the post-hoc procedure from Conover for a Friedman test. Unfortunately I do not have a copy of the original book 'Practical Nonparametric Statistics' and it seems a bit expensive. So I've been googling and came across two different formula's both claiming to have Conover as a source, and both give different results.

The first formula is shown in the documentation on the PMCMR package on p. 17. The documentation can be found here. The same formula is also shown in a book 'Experimental Methods for the Analysis of Optimization Algorithms' on p. 319 (google books has this page, hope this link works).

Then there is another formula that produces different results shown on this website, and is also used in the SPSS syntax created by Timo Gnambs available here.

I followed both formula's in Excel and they ended with quite some big differences.

My main concern is the following: Did Conover simply discuss two different methods in his book, or is one of the two formula's wrong?

Thanks in advance for any replies

p.s. I tried to show the formula's here with LaTeX but unfortunately I couldn't get it to work. Also the search feature on the site is not working.