Also the upper of the two antennas in P-30 is actually tilted by 25 degrees:
http://www.radartutorial.eu/19.kartei/karte921.en.html
Hmm.. I didn't notice that. So that is a scanning of two beams and location by a direct 3D solver?
A truncated oval antenna like the ones you can see on all these pictures generates a radar beam which is a narrow stripe perpendicular to the long axis of the oval. This means that the horizontal antenna of the P-30 creates a vertical radar beam, while the tilted antenna generates a beam stripe which is tilted 25 degrees off the vertical axis.
Now both antennas are mounted on the same motor axis and rotate. At ground level both the vertical and tilted beam are parallel and any reflection from a target at ground level will show up at the same time for both antennas (which are working on different frequencies or sending out pulses with a time delay). But a target which appears at a higher elevation angle from the radar site, the tilted antenna will see it when the horizontal antenna is already pointing in a different direction - the higher the elevation, the further the upper antenna has to be rotated to "see" this target.
From the time it takes for the echos to return to the antennas you get the distance to the target and then it is just pure trigonometry to also determine the height of the object above ground.
The newer systems with two horizontal antennas are working simultaneously on several different frequencies or using a frequency sweep together with an antenna system which directs the different frequencies into different elevation angles. From the echos at the different frequencies the system then determines the elevation angle and from the time delay the distance.