Secure circuits are prone to a wide range of physical attacks. Among them, fault attacks are based on modifying the circuit environment in order to change its behaviour or to induce faults into its computations. As a result, the security level of the circuit under attack may be weaken. Many means are of common use to inject such faults: laser shot, electromagnetic pulse, overclocking, chip underpowering, temperature increase, etc. However, the mechanisms involved in the fault injection process have not been yet deeply investigate. Especially, those that have a global effect linked to timing constraints violation. In this paper we provide an experimental proof of the uniqueness of the fault injection process by means of the target's clock, power supply, or temperature alteration. We also studied further the properties of these fault injection means. These insights are intended to give designers guidelines to strengthen fault countermeasures. It also enable to imagine broad-spectrum countermeasures against most of the fault injection means.