\documentclass{standalone}
\usepackage{tikz}
\begin{document}
\tikzset{
partial ellipse/.style args={#1:#2:#3}{
insert path={+ (#1:#3) arc (#1:#2:#3)}
}
}
% copied from: https://tex.stackexchange.com/questions/123158/tikz-using-the-ellipse-command-with-a-start-and-end-angle-instead-of-an-arc
\begin{tikzpicture}
\newcommand{\wa}{7} % angle 1
\newcommand{\wb}{42} % angle 2
\newcommand{\Ra}{1} % radius 1
\newcommand{\Rb}{2} % radius 2
\newcommand{\B}{5} % width
\draw[rotate=\wa] (0,0) ellipse ({cos(\wb)*\Ra} and \Ra); % draw the small
\begin{scope}[rotate around={\wa:(({cos(\wb)*\B},{sin(\wb)*\B}))}] %fill=gray!30,opacity=0.2
\draw[] ({cos(\wb)*\B},{sin(\wb)*\B}) [partial ellipse=-90+2*\wa:90+\wb-2*\wa:{cos(\wb)*\Ra} and \Ra]; % draw half of the smaller ellipse at the rear side
\draw[dashed] ({cos(\wb)*\B},{sin(\wb)*\B}) [partial ellipse=90+\wb-2*\wa:270+2*\wa:{cos(\wb)*\Ra} and \Ra]; % draw half of the smaller ellipse at the rear side (dashed)
\end{scope}
\draw[dashed] (0,0)--({cos(\wb)*\B},{sin(\wb)*\B}); % center axis
% wall
\draw ({0-sin(\wb-2*\wa)*\Ra},{0+cos(\wb-\wa)*\Ra}) -- ({cos(\wb)*\B-sin(\wb-2*\wa)*\Ra},{sin(\wb)*\B+cos(\wb-\wa)*\Ra});
\draw ({0+sin(\wb-2*\wa)*\Ra},{0-cos(\wb-\wa)*\Ra}) -- ({cos(\wb)*\B+sin(\wb-2*\wa)*\Ra},{sin(\wb)*\B-cos(\wb-\wa)*\Ra});
\draw[->] (-0.2,0) -- (4.2,-0.5119) node[right] {x};
\draw[->] (0,-0.2) -- (0,4.2) node[above] {y};
\draw[->] (0,0) -- ({-sin(\wb+\wa)},{-cos(\wb)}) node[below] {z};
\end{tikzpicture}
\end{document}
Hier ist eine Möglichkeit den Zylinder einzufäben und zu schattieren.
Eigentliches Ziel des Links aus dem Kommentar von @Murmeltier: https://texwelt.de/wissen/fragen/22898/was-ist-der-unterschied-zwischen-tikz-und-pgf/22917