Since you seem to have a scope with sufficient bandwidth, do you have a generator capable of generating a square wave of several hundred kHz?

If you have one, you can measure ft using the following procedure with a relatively simple circuit.

View attachment 1340776
1. set the collector current by adjusting VR1.

2. apply a square wave to the input (probe 1) and adjust the value of R2 (actually a trim pot) so that the output waveform (probe 2) is a square wave - a similar procedure to adjusting the probe on a 10:1 oscilloscope.

3. read the input amplitude Vin and output amplitude Vo on the scope.

4. determine Ft using the following equation.

Ft = (1/(2*π

*C1*R1))*(Vo/Vin)

Note that there is no "Ft" in the model parameters in SPICE, but the forward transition time "TF" is specified.

Since TF = 1/(2*π*Ft)

The TF can be calculated as follows.

TF = C1*R1*(Vin/Vo)

Specific steps are shown on the simulation using the 2N2222 model as DUT.

1. aiming for a measured collector current of 20 mA, VR1 was adjusted to 20 mA at the 27.7% position. In the actual measurement, measure the voltage at both ends of R1 with a DMM.

2. add a 500 kHz 100 mVpp square wave to the input. Adjust the value of R2 while checking the output waveform: if R2 is 6 kΩ, the rising and falling edges overshoot;

View attachment 1340777
if R2 is 4.5 kΩ, the rising and falling edges round off;

View attachment 1340778
if R2 is 5.23 kΩ, the output waveform is a clean square wave.

View attachment 1340779
3. The output amplitude at this time was 78.6 mVpp.

4. Ft = (1/(2*π*

*20p*20))*(78.6m/100m) = 312.7MHz.

TF = 20p*20*(100m/78.6m) = 0.509nsec.

I checked the parameters for my 2N2222 model and found TF = 4.00E-8 i.e. 0.4nsec.

This difference could be reduced by increasing the collector current value a little more. However, be aware of the heat generated by the DUT in the actual measurement.

I built and used a board for this measurement almost 20 years ago. The left one is for NPN and the right one for PNP. The traces on the boards are exactly the same, only the polarity of the power supply and electrolytic capacitors is different.

View attachment 1340780
View attachment 1340781
(Some differences from the circuit diagram above)

R1 uses a plate resistor with low inductance, but SMD is better. Due to the availability of parts on hand at the time of production, 20pF seems to use two 10pF SMDs. (It was a long time ago, so I don't have a very accurate memory.) If it is a circuit of this level, I think it is possible to produce it with a perfboard.