Hello! I am trying to measure the peak of a 1ns pulse at 1MHz coming from a photodiode using Arduino. I know that Arduino is not fast enough to measure this directly, but I was wondering if anyone had any suggestions on how to properly stretch the pulse to a duration that perhaps the Arduino could read, and/or any other detection strategies. I am trying to measure this as part of an optimization scheme to maximize the peak voltage of the pulse. I don't need to read the exact voltage; I just need to be able to tell if it is optimized so as long as there is some sort of proportionality to the peak amplitude, that would work. Typically the pulse amplitudes are quite small, around 5mV, so I would also need to amplify a bit. Thanks in advance for your help!!
-
A One-shot can be used, however, a 1ns pulse will be a problem.
-
Create an image of the signal being monitored and use it to communicate what is needed/being done; similar to an oscilloscope display.
Really? With a 1ns pulse you are far in the GHz frequency region. Rise and fall times are in pico-seconds region. To measure such pulses requires very expensive measurement equipment and a measurement setup for the GHz range. A 'normal' measurement setup will already greatly influence and change such a pulse. You will not measure any realistic values, but only crap
To do that accurately, you will need an ADC that can sample at 10 GHz or higher.
Most people instead use a simple RC circuit to integrate the pulse, and capture the integrated peak at 1 MHz or less.
There is this Arduino/Silicon Photomultiplier gamma spectrophotometer project that may give you some ideas.
Another resource: theremino | the real modular input output
Photodiodes output a current unless you are already converting the current to a voltage.
So what other circuitry is involved in your set-up?
There is an old logic family that may work called ECL (Emitter Coupled Logic) made by Motorola and a few others. It was fast, possibly fast enough to do what you want. There are some RF parts that may work as well. look at microwave components this is a 1 GHz signal.
I'm not an expert, but maybe a peak detector; a fast opamp (like OPA2677 or faster), followed by a diode to charge a capacitor. Then a resistor to discharge the cap and a (fast) comparator. Or maybe the capacitor will self discharge without help.
The small amplification, capacitor and discharge resistor should be calculated to stretch the pulse time a bit. I don't know if this could work for such small pulse.
I was also thinking of a peak detector. I use them "all the time" for audio.
But... The speed is likely be a problem. Op-amps have speed/frequency limits and it takes some time to charge-up the capacitor. A transistor to boost the op-amp's output current could help to charge the capacitor faster but I've never tried that. If the pulse is consistent and repeated, the capacitor will get more fully-charged with each pulse so that will help.
Obviously, the photodiode amplifier also needs to be "fast".
I'm not an RF expert but things get tricky at high frequencies! (Impedance matching, SWR, return loss.)
I'd say that is what an oscilloscope ( of the proper bandwidth ) does.
If you need to automate something you can define a pass/fail mask ( in the scope ).