In the x-z coordinate system, the angle between the direction of proton emission and positive z-axis is 80°. The radius of the source is 0.6 cm. The projection in the x-y coordinates is going to be an ellipse. However, the shape of the energy deposit calculated by USRBIN is a circle, which should be a ellipse. The input file and picture are below:

Dear Cerutti:
Thank you very much.
But, in x-y plane, the shape of beam should be ellipse, not cycle.
I think, the source setting parameters have some mistake. May be BEAMAXES should be added. But, there are some problem. Please look at the picture in the next reply.

Why? You defined an annular beam in the x-y plane and this is what you get. Then, you have defined the beam direction at 80 deg in the x-z plane, therefore the circular annulus moves (not much, due to the low energy) along it, but the x-y shape remains circular as it should (as discussed on slide 15).
If you meant instead to get a circular beam shape in the plane perpendicular to the beam direction (that is not xy), such as to reflect into an ellipse on xy you expect, you should indeed make proper use of BEAMAXES as indicated by David.

Dear horvathd,
Yes, you are right.
But, in the inputfile, the length of what(5) and what(6) is 10 bytes. If decimal numbers less than 10 bytes, there is also a mistake.

The BEAMAXES card is added. But the results is still not right. Look at the next picture, when used the BEAMAXES, the radius of the beam become less than 0.6 cm. So, what’s the problem about this phenomenon?

the issue is that you set the beam direction along the Y axis on the BEAMPOS card. So, the particles will propagate sideways from the starting surface.

When you are using BEAMAXES and you want to keep the beam direction perpendicular to the beam’s shape, they you should not change the beam direction on the BEAMPOS card.