Tube amp improvements: some measurements (part 4)

After assembly the tube audio amplifier, I have got some measurements to plot the frequency response magnitude.
I have used a 8ohms purely restrictive load (figure below).

I swept the frequency of an sinusoid input signal from 20Hz to 20kHz. I took the outputs signals in an oscilloscope Tektronix Type 422.
Bass and treble adjust were on the maximum position.
Figure below shows the frequency response magnitude.

I haven't noted any distortion. Even in low frequencies, the output sinusoid is free of distortion.
Figures below shows outputs at 20Hz, 30Hz, 50Hz and 80Hz.

Following my propose (previously posted: tube amp improvements part 1):

1) Big output transformers (high inductance) provide a good response at low frequencies and without distortions.
2) I can't evaluate yet the benefits of bigger power source capacitors. I have not seen ripple at output. I can hear a humm noise but the circuit in not shielded yet. After I conclude the shield I will can evaluated the benefits of bigger capacitors.
3) I tried bigger coupling capacitors and also eliminated them in some connections, but I have not noted significant results for the bandwidth.
4) The next step will be: tests with output transformers at ultralinear configuration. I will connect the screen resistors to the specific wires of the output transformers primaries.

My first impressions: very good output transformers and tubes (Sylvania).