Samfish, do you have the emission spectrum of your light? Or do you have details of which brand/models of LEDs are used at what proportion? If so, you can calculate the conversion factor (between lux/fc and PPFD). I learned it from this thread:
Math behind | Rollitup
It took for a while to figure it out, but it is pretty cool once you learn it. If you have the figure of emission spectrum (e.g. in forms of spectrum power distribution, SPD), you can digitize it with something like this:
WebPlotDigitizer - Extract data from plots, images, and maps
I think someone or I posted about a cheap USB PAR sensor from BioTek Marine (based on Apogee sensor), but now Apogee released their own.
SQ-420 Smart Quantum Sensor | USB Output - Apogee Instruments, Inc.
$195, and the cool thing about it is that it can log over time intervals. The logging is irrelevant for artificial light, but it is useful for greenhouse growers to get the daily PAR (called Day Light Integrals, DLI).
But as Terry mentioned, Apogee would underestimate deep red (680nm). But if the emission spectrum is known, the correction factor can be calculated for anything (including 680nm LEDs) in the similar method as the Rollitup post I linked.
As Terry mentioned, with constant light, you can get away with fairly low light. Just put it into a context, 500fc of sun light contains 100 micromol/m^2/s. I use a bit more than Terry, though. I've been doing 100-150 micro mol/m^2/s, 13h/day (for Paph/Phal). Maybe I can go even lower. How long is your day length, Terry?