My solution involved making my own regulator from bits I had lying around I'm afraid. Picaxe 18 pin chip, i2c LCD panel and some Mosfets.
I certainly don't think you can combine the inputs of solar and wind onto the one input on the Tristar, that'll buggers something up for sure. Both have VERY different voltages at a guess, plus the windgen is a rectified unsmoothed AC wave form, while the panels
are pretty smooth DC.
I think you face the same problem I faced, which is that the diversion load is on "the wrong side" of the regulator, to my mind. It is not a true "diversion" the way it is pictured in the tristar manual, but a drain load, and that is going to play havoc with the charging
I just programmed the picaxe to divert power from the Aerogen
to the dump load once the batteries
reached 13.6 volts, and left the whole top end of the battery charge algorithm to the Vitronic reg on the solar panels
. Have not yet actually tested it in anger because the boat
has been in the pen since dinosaurs roamed the earth, but it worked perfectly on the garage bench with simulated inputs from a variable power supply. I only fitted the LCD to give me some sense of how much power I was getting over time from the Aerogen, one of those toys where the novelty will wear off pretty quickly I suspect. (We call it EFS, or Executive Fascination System, in my industry)
When I have time to play around, I will see if I can use a memory chip on the I2C bus of the picaxe to spot the final stage charging voltages from the Vitronic and contribute a bit of power if needed, but that is a lot of effort for little gain I suspect.
Sorry I could not come up with a better solution.