Photo-flash duration and trigger lag measurements
Although a photo-flash fires much faster than a camera, the timing might
become critical for high-speed photography. Below you will find the flash
duration and the trigger lag numbers for the popular Canon Speedlite 430EX
To interpret the oscilloscope screen-shots it's important to understand how a modern photo flash unit controls the output power. Instead of wasting one part of the energy to reduce the light output in modern devices like the 430EX the discharge is interrupted by a transistor (IGBT). The remaining energy in the photo flash capacitors can be used for the next flash reducing the charge time. Only in the 1/1 output setting the whole energy in the caps is discharged.
Let's have a look on the flash duration diagrams. They were taken with a simple photo diode (SFH208P) loaded with a resistor of around 500 Ohm for a fast response. The amplitudes are not of importance here. The scope was triggered by the falling edge of the trigger voltage after short-circuiting the trigger pin of the flash unit.
For convenience I overlaid all seven plots for 1/1, 1/2, 1/4, 1/8, 1/16, 1/32 and 1/64 in one image. The time base is 500 us/div.
Below is the same measurement with a time base of 100 us/div allowing a closer look at the shortest times.
Finally the flash duration for the lower limit of 1/64 of the Speedlite 430EX is shown.
The energy output can of the flash can be interpreted roughly as the area below the curves. Correspondingly the span of the flash durations is very large ranging from 2 ms to 70 us. This results in a factor of about 30. The flash durations are listed in the following table. A threshold of about 1/5 of the maximum amplitude was used.
|Power setting||Flash duration|
|1/1||2000 us = 1/500 s|
|1/2||750 us = 1/1333 s|
|1/4||330 us = 1/3000 s|
|1/8||240 us = 1/4200 s|
|1/16||160 us = 1/6200 s|
|1/32||120 us = 1/8500 s|
|1/64||70 us = 1/15000 s|
The trigger lag is the delay between the first trigger pulse edge and the moment when the light output of the flash rises. This lag is usually negligible for higher power settings, but becomes increasingly important for shorter flash durations. It's value might turn out critical especially when using more flash units. The following scope screen-shot shows the ignition phase of the Speedlite 430EX.
The trigger lag is about 50 us. About 32 us after the trigger pulse edge a spike
in the measured signal can be observed. A possible explanation might be the
ignition transformer coupling in. Following this assumption the lag would
be distributed with 32 us on the input circuitry and about 20 us on the ignition
Below is the same measurement for the bigger Speedlite 580EX2.
The result here is about 30 us and noticeably shorter then for the 430EX. Here the spikes are followed directly by the rising light output. This measurement was made in a different session than the upper one so the boundary conditions might have changed.
Similar measurements for a Sigma flash made with a more sophisticated sensor.