I have the following circuit that uses a NPN transistor as a switch:

Circuit schematic

I've built the circuit on a PCB, and while testing I noticed the following plateau in Vce (blue trace) when Vb transitions to zero (yellow trace):

Scope traces

Does anyone know what is causing this? It never showed up in any of my simulations with LTSpice.

I've played around with including a BAT54 Schottky diode between base and collector to prevent saturation, but the odd plateau remains.


That's mostly the effect of Ccb. The collector-base capacitance.

During the transition region Q1 is acting as an amplifier and the quick rise of the collector biases the base through Ccb. This slows down the transition until Ccb is sufficiently charged.

During the transition you can also see Ccb as a large Miller-multiplied capacitor from the base to ground, slowing the response of your base-drive network,

Although some call this the Miller effect as this plateau can be seen a consequence of the Miller effect as captured in the Miller theorem. The Miller effect refers to linear systems and it only applies to the narrow window during which the transistor is acting as an amplifier, this allows for circuit simplifications. It says nothing about switching systems entering and leaving such linear region, although it contributes somewhat to their understanding and problem mitigation.

  • \$\begingroup\$ Do you know of any methods to combat the effects of Ccb? \$\endgroup\$ – user181297 Dec 5 '18 at 21:30
  • \$\begingroup\$ @user181297 increasing base drive is the obvious one. That is the main purpose of C1. It creates a capacitive divider with Ccb and the collector. Increasing that capacitor, and perhaps adding a diode to avoid the large negative transient that imposes, would work to some degree. But you will quickly hit diminishing returns. \$\endgroup\$ – Edgar Brown Dec 5 '18 at 21:39
  • \$\begingroup\$ A cascode amplifier will not suffer from the miller effect, but has other downsides. If you get desperate you could try that \$\endgroup\$ – C_Elegans Dec 6 '18 at 1:25
  • \$\begingroup\$ This is Miller Effect. The delta_charge on the collector is exactly the delta_charge on the base. The delta_voltages end up being negative feedback, where the delta_collector fights the delta_base. Without the Cob charge feedback, the turnoff time would be faster. \$\endgroup\$ – analogsystemsrf Dec 6 '18 at 3:22
  • \$\begingroup\$ @analogsystemsrf what is the Cob charge feedback? Is this discharging or charging of the output capacitance? \$\endgroup\$ – user181297 Dec 6 '18 at 16:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.