Good point Peter.
Luckily Ha and SII emission lines are both red and thus have similar wavelengths, and OIII is only about half way between red and blue, so we don't really need blue in a classic HaOIIISII narrowband imaging.
I had absolutely no registration issues with an affordable doublet when imaging at f/5.6 ;however, halos around bright stars were noticeable due to the out of band signal getting through.
Here is another example of a tricolour narrowband semi-decent image taken with the previously mentioned doublet:
https://www.spacetelescope.org/proje...r_lipinski_06/