Vertical Curve - Parabolic Definition and Equivalent Radius

All

I am working on a spreadsheet to check designs. I am essentially looking to input vertical intersection points and curve lengths, after which the speradsheet will then output gradients, gradient lengths, and curve radii. 

BRT is set to vertical curves as parabolic and annotated with equivalent radius R = 100K

My issue is, that the vertical radii I calculate is different that what BRT presents. My calc is R= 100*L/G where L is length of curve and G is the difference in gradient. 

For two grades of G1=0.638% to G2=0.589% and L =30m I get R=61124.4898m

BRT gives R = 60653.170m

Now I am willing to give BRT the benefit of the doubt over a couple of m due to accuracy in the input compared to what the software holds, as well as the conversion from K to equivalent R, but this is a ~570m difference in the calcs.

Can anyone advise where this is going wrong?

Thanks

A