hi, I had a problem about the confined sources, and I hope someone know how to solve it. I need to use some hot rods phantom like the Derenzo phantom or the Micro Deluxe phantom. I know there's a simple way to define those phantom by using the "quadrant repeater" in the phantom and later confine a big source to this rod. When I tried this way, run a short simulation and bin the of the sources' coordinates, I found that one source is different with the others, the one in the up-right corner is not totally filled. I've tried different phantom sizes, there were always some rods distribution not totally filled. From the attachment you can see another more complicated phantom with the same phenomena. Because I don't know how to fix this problem, so what I did was to define all the rods  manually as the sources and not using the "confine" methods. In this way, the results come out correctly. However, in this way, we need to calculate all the coordinates of the rods and put it into the source's .mac file. Now I want to run a simulation with one phantom in different locations. I can set the center positions of the phantom in the *.mac file as parameters, however, I cannot set all the positions of the sources in the same way, it's not practical to set hundreds of parameters of the rods' location as the input parameters of Gate. Therefore, I want to go back to the original idea of using confined sources. Has there someone else noticed this problem about the confined sources? And how you handle this? Please give me your advise. Thank you. Yuxuan