Use the sine-bar bandwidth pattern:
Engrave it in grayscale mode as a negative image with 0.1 mm line spacing:
Monitor the Ruida KT332N controller’s analog laser power control output:
- 1 X axis
DIR, low = left-to-right (yellow)
L-ONlaser enable, low active (magenta)
L-ANanalog voltage (cyan)
The scope triggers when the top two traces go low during a left-to-right scan with the laser beam active. The trigger point lies far off-screen to the left, with the delay set to pull the interesting part of the scan into view.
Although both the controller’s
L-AN output and the laser’s
IN input specify a signal range of 0 V to 5 V, the analog output voltage never goes below 0.4 V, but (as will seen later) that produces 0 mA from the laser power supply.
Set the X cursors to the top and bottom of the sine wave and read off the 4.36 V peak-to-peak value.
Set the Y cursors to matching points on successive cycles and read off ΔT=33.44 ms. Because each cycle is 1 mm wide, the scan speed is set to 25 mm/s and traveling 1 mm should require 40 ms, puzzle over that number and the related fact that 1/ΔT=29.91 Hz. This seems to happen only for speeds under 50-ish mm/s, for which I have no explanation.
Repeat the exercise at various speeds up through 500 mm/s:
The analog output voltage has dropped to 1.56 Vpp.
The average voltage increases from 2.66 V at 25 (or is it 33?) Hz to 2.78 at 500 Hz, which is reasonably close to the same value.
The signal’s -3dB point would be at √½ × 4.36 Vpp = 3.1 Vpp, which happens at 200 mm/s = 200 Hz:
Which is eerily close to the “around 200 Hz” bandwidth figured from the risetime measurements.
All of the analog output measurements as a slide show:
One might now wonder whether there’s any bandwidth difference between the analog and PWM signals as measured in the laser tube current.
Data! We need more data!