Issue with small reference contigs

[Csam488]

When a small (2-20kb) reference contig is used it appears that a significant portion of reads generated are the complete reference genome and have the exact same sequence (IGV trace attached for context). I have tested this with both provided pore models and various -e ,-f, and -s values. I unfortunately don't know if this is DeepSimulator functioning as intended but it's causing difficulty as I'm attempting to simulate sequencing of PCR products. Could this potentially be solved by introducing a maximum read length parameter?