need help with project measuring frequency response for headphones..

[vk6zgo]

I used to do something like this for a living when I worked for the Hearing Aid place.
We used to do tests on  Audiometers,using an artificial ear which was calibrated each year at an Acoustic Research establishment.

You had to very carefully position the headphones so that the test would give correct results.

[olsenn]

Unfortunately, headphone measurements are not very objective. Even if you did buy a multi-thousand dollar dummy head and expensive microphones with low distortion and a known frequency response, then the measured results would still not come close to what you perceive. Sites like Innerfidelity and headroom use a dummy head and then apply what's called a "Head Related Transfer Function" (HRTF) to manipulate the graph to look more like what you hear... but this transfer function is just made up on the spot by audiophools (aka not objective). Typically the first kilohertz will be fairly accurate without the need for a HRTF, but after that it's a big mess.

[Lightages]

Exactly what I have been trying to say. Headphones are subjective and always will be. You can make a dummy head of your own head and ears and have a good approximation of what you are going to hear but it is only for that person.

However, the OP is asking how to measure, not whether he should or not, so we should give him the information on how to measure, not tell him he shouldn't try.