Yes I always wanted to do them. For the original set I just compared them to my friend's expensive headphones. It was only today that I did any kind of frequency response testing. But since I don't have a calibrated microphone, it is more useful for doing relative comparison between my headphone designs than it is for providing absolute numbers.

What I wanted to do was build a little headphone test dummy head. But for my relative tests I just shoved my podcast mic, Audio Technica AT2005USB up to the center of the driver on one side of the headphone. Do look at my linked tweet for those FR curves.

However this little adventure has encouraged me to consider building the little head dummy. I have some nice little capsule mics I was playing with that would work nicely, and I have a datasheet for those that probably includes a curve.

This hobby work of mine goes slowly, but I appreciate your suggestion. I will try to tune up my FDM 50mm headphone design so that it sounds at least as good as my original resin printed design, and then look at building a better characterization rig.