All spectra I've been able to find are taken above 400 nm. This is probably due to the fact that a spectrograph + ICCD is needed to be able to do this.
Actually I never tried to do this but if I consider as a source the flame of real wax candle, the issue is that my solid state spectroradiometer has a sensitivity in the range 200-300 which is very lower in comparison with the range 400-1000
Since the spectrum is expected to be next to a black body I think that no useful information in the UV wavelenght would be gathered.
Next time I try a measure I will report but why do you need to measure a candle?
For an experiment we are using a mercury filter 257nm on a intensified high speed camera. by incident we noticed that we could see the flame of a lighter. now we use a candle to set the focus. However the highest intensity we are measuring is in the yellow part of the flame. So we where wondering if it is UV that we are measuring or if the filter leaks in the visible range.
I wonder instead if the filter leaks more easily in the near IR; anyway it seems to me normal that you measure the highest intensity around the yellow because the detector has possbly the highest sensitivity in that range and the flame has actually a relevant yellow emission, even if lower compared with NIR