I am sure this varies given the distance from moon to earth varies, but a range would be sufficient. I am trying to explain to a flat earther how there is not a lunar eclipse every full moon.
My approach is to show that the 5 percent inclination of the moon's orbit to the ecliptic will result in a large enough gap for the umbra to miss the moon when projected at a great distance.
The angle of the cone is due to the angular extent of the light source (the sun). This is approximately 0.53 degrees when near the orbit of the earth.
So the length of this cone is: $$ \tan\left(\frac{\theta}{2}\right) = \frac{R_{earth}}{L}$$ $$L = \frac{R_{earth}}{\tan\left(\frac{\theta}{2}\right)} $$ $$L = 1.4\times 10^9 \text{m}$$
The diameter of the umbra varies linearly as your displacement from the vertex. The moon at perigee is about $0.36\times 10^9\text{m}$ from earth or about $1.0\times 10^9\text{m}$ from the vertex.
Therefore the diameter of the umbra at that distance is approximately $\frac{1.0}{1.4} D_{earth}$ or about $9100\text{km}$. Smaller than that when the moon is further away.
