you mean like the iBcrazy Skyhammer? The first pic here is the general pattern (the remainder of the post revolves around mad mushroom v. skew planar antennas . . . not sure why he posted that pattern, but it's a handy plot for understanding the shape).
It doesn't exactly work the way that you think it would.
RF theory gets generally weird in these cases and treating two antennas that have been joined by a junction in an network is not the same as two independent antennas in a diversity receiver. With the skyhammer setup you end up with a stronger horizon lobe at the horizon, but instead of a point null at zenith and nadir you end up with two circular nulls 45 degrees off the Z axis. if you're flying genuinely on the horizon and your antennas are fixed on the base station, not a problem. If you might wander into that null (flying close by or tilting your head on a head-mounted antenna), you're asking for the signal to fade at random to nothing.
This assumes the antennas in the network are matched polarity. If they're mismatched . . . Just don't. Even on a diversity setup, it's not a good idea. The whole point of CP is to reject multipath noise by discarding the mirrored polarity from reflections. Add a reverse polarity antenna and you've just added them back in. With a diversity RX at least the RX will filter out the weaker of the two, but if reflected signals ever become stronger than the main beam, a: you've done something bad (probably obstacle masking) and b: don't expect the reflections to be pretty.
As far as patterns for a network of CP antennas orthogonal to each other . . . I don't want to even contemplate the ugly that that pattern would generate. It would be an interesting assignment for an RF class, but a horrible antenna.
makes sense, yeah wasnt expecting it to be like 2 independent antennas, since the voltage would also already be split, but thats funny sort of like putting 2 grapes in a machine and getting 1 raisin on the output