Hi
I have done search of this thread and no one seems to have mentioned latency.
I have a test setup using a LM393 running on a particle photon with a sketch that also runs a flashing led. I have tested this with various led flash lengths and intervals, the highest rate being a 10ms flash with 10ms interval . I have a consistent precise reading of pulse interval and number but have identified that there is a 1ms delay between the flash and the start of the interrupt function. This latency is consistent, whatever the interval between and I can only assume it is a result of latency caused by LM393 and interrupt, though I find it strange that the latency is exactly 1millisecond even when measured in micros. Under low meter loads this discrepancy would most likely go unnoticed but with high loads of 72kWs it results in a significant error when calculating the instant power.
Has anyone else noticed this or can give another reason for this that I am missing?
A subtraction of the identified latency from the pulse interval reading, calibrates the setup and produces exact readings