Ground Plane Measurements: Determining delay / offset

[akunec]

I just made some ground plane measurements of my speakers mounted in their cabinets with a 42ms time window. In order to calculate the acoustic center offset between the drivers I did the following:

Mounted microphone at 2m on tweeters axis.

Measured tweeters response and calculated its phase using HBT.

Measured mid's response (and shifted volume until its about the same as the tweeter). Determined phase using HBT.

Measured tweeter and mids response together.

My understanding is that when I tested them together, it would tell me of how much they are in phase with each other. So I imported the mid data and tweeter data into LSPcad and put a delay on the tweeter. I adjusted the delay (using optimizer) until I reached the transfer function of the combined mid and tweeter response.

I achieved a delay of only 5.13 microseconds, which really doesn't seem right. Can anybody tell me why this didnt work?