I have the standard lense that it comes with when purchased. Tank is 26 inch deep and lights are about 7-8 inches off of water.
The
white graph looks good to me for a deep tank and you will probably want to get eventually to 100% for what I assume, by your name, is an acro dominated system,
but everything else looks off in relation to the white. Also, be really careful directly under the clusters in the top 1/3 of the tank since the PAR is obscene and you will have a nasty hotspot (but this can be good for some super high light loving acros like Millies, Abros, Efflos, etc) you need to be aware of.
Your blue looks very, very low compared to your white and other colors. You should have your blue set at the same level as your white throughout the day for growth and color. While white is great for acros, so is blue - you will generate more PAR with your blue lights than with your white lights. When you boost your blue, you will have to drop your white for a little since the total intensity is going to go way up, but you can eventually get them back to where they are in a couple weeks.
The spectrum graph on the bottom is your friend - especially if you want to emulate the growth you had with T5s or MHs.
The best thing I did for my tank and my Radion configuration was borrowing a PAR meter. I now have PAR graphs for my entire tank in various different settings. I don't use the default lens (I have the wide-angles) so I can't give you comparable numbers - but you are heading in the right direction.
Your red and green levels would be my only worry since a lot of red light, especially in the hotspot, isn't going to be good. But, I can't see their relative intensities so they might be fine.
If you want a custom light graph, you really need a PAR meter otherwise you are just shooting in the dark and will probably **** off your tank. Also, always start off with a tried and true spectrum by copying a nice T5 configuration or a quality MH bulb and then slowly adjust things from there to maximize growth and color. Don't start with some whacky spectrum with unknown PAR values and expect anything.