I need to log data from an aparatus that generates a voltage output as an instrument sweeps through 360 degrees. Logging just 1 revolution is sufficient although the equipment runs until it's turned off! (However, the more revs I can log to compare with each other and or average, the better!). The output is - or can be scaled to be within - the 0 to 2.5V specification.
The instrument will run at around 20,000 RPM, which equates to approx 333Hz. To make the data useful I need to sample every 20 degrees, minimum. Every 10 deg would be better and every 5 deg better still.
This equates to taking(360/20) x 333Hz = 5,994 readings / sec at 20 deg steps, or 11,988 / at 10 deg steps, or 23,076 at 5 deg steps.
There will be upto (at this point) 8 chans of data to log. One for a once-round-the-circle datum pulse (of 2.5V from a monostable). Another for a degrees stepped pulse (another 2.5V pulse from a monostable -say- at 20 / sec) positionally locked to the datum pulse. 6 more chans will read the data. The idea is to relate the pulses to the data chans to see what is happening at various parts of the revolution.
The pulse reading chans are well within the range of the loggers' 1kS/s. But the data chans clearly aren't! I cannot slow the rotational speed to take this into account.
I am unsure whether / not the logging speeds will handle my data sampling needs in any of the sampling modes of the loggers' operation.
{For example it might be OK to run the whole test doing it just one data chan and the two pulse chans at a time. I would then be have to run this 3 chan log as many times as there are data chans to read - say the 6 times previously mentioned. Clearly, though, I have to be able to save the data somewhere and identify it for later processing, and begin logging again with the next test, and so on.}
Is there any way your logger can meet my needs? I'm very flexible in the way I can run the tests so I'm hoping that, somehow, there is some way this can be done with a Picolog.
I've no chance at writing software. I know nothing about it at all so the SDK option is, effectively, no option for me. And I don't suppose I'll get anyone to write it for me!
I'll contact your support guys and see what transpires, tho.