Yeah, you can take some photodiodes that far. It's a bit of a PITA to generate -30 V, and extrapolating from the -10 V experiment, I suspect it still doesn't fix it entirely.
How do you know the LEDs are generating a clean square wave light output when you pulse them for 10 us? I've tried to do something similar to what you're doing (completely different application) and wasn't sure my LEDs were generating clean pulses - if not, the messy photodiode output might be "real". I have been planning to setup a mechanical light chopper (spinning slotted disc) to generate light pulses with known edge characteristics. (Quite possibly you know more about LED physics than I do...)
For one, increasing the reverse bias of the photodiode markedly reduces the effect. Plus, I experimented with other small LEDs (red, blue) and didn't see any real difference.
I'm not sure what mechanism could cause a LED to keep glowing on its own. Capacitive and inductive effects should be tiny. Some white LEDs used for household lighting may use phosphors to improve spectrum, but small single-color LEDs should be fine.
I've been driving the LED with the output from a signal generator (at MHz rates). I saw the same sloppy photodiode output you've documented, despite using a optical detector module (integrated opamp) supposedly rated for 1 GHz. So I began suspecting the light source. Capacitance in the driver circuit? Probably it's just the effect you've noted. I'll be watching your progress here with interest!
I seem to recall -30V reverse bias in a datasheet, maybe the analog devices one you referred to earlier.
Yeah, you can take some photodiodes that far. It's a bit of a PITA to generate -30 V, and extrapolating from the -10 V experiment, I suspect it still doesn't fix it entirely.
How do you know the LEDs are generating a clean square wave light output when you pulse them for 10 us? I've tried to do something similar to what you're doing (completely different application) and wasn't sure my LEDs were generating clean pulses - if not, the messy photodiode output might be "real". I have been planning to setup a mechanical light chopper (spinning slotted disc) to generate light pulses with known edge characteristics. (Quite possibly you know more about LED physics than I do...)
For one, increasing the reverse bias of the photodiode markedly reduces the effect. Plus, I experimented with other small LEDs (red, blue) and didn't see any real difference.
I'm not sure what mechanism could cause a LED to keep glowing on its own. Capacitive and inductive effects should be tiny. Some white LEDs used for household lighting may use phosphors to improve spectrum, but small single-color LEDs should be fine.
I've been driving the LED with the output from a signal generator (at MHz rates). I saw the same sloppy photodiode output you've documented, despite using a optical detector module (integrated opamp) supposedly rated for 1 GHz. So I began suspecting the light source. Capacitance in the driver circuit? Probably it's just the effect you've noted. I'll be watching your progress here with interest!
I'm fascinated what the project might be now