If a non grounded piece of metal is subjected to an ESD, a spark between this metal part and grounded metal parts can occur. This discharge is generally called secondary ESD. These secondary ESDs are often very close to the electronics and, as this article shows, can have much higher currents and shorter rise times than the original ESD. For that reason secondary ESD poses a very high risk of causing soft-and hard errors in the affected Device Under Test (DUT). A methodology to model the secondary Electrostatic Discharge (ESD) inside a portable electronic product is presented. It is a hybrid method that combines linear descriptions of the ESD generator and the DUT with the nonlinear spark model and a model for the initiation delay (statistical time lag) of the spark. Measurement results are presented comparing discharge currents and time delays for two cases: secondary discharge between metal rod and ESD target, and inside a portable product.