I think the difference could be in the way that you are treating the shot 10 timestamps.
As shown by the color coding in my timestamp table, I was not able to discern boom times from the audio data for shot 10, except for TMX (Source 2 - fence).
For the other six sources, I used the crack-boom difference formula that you derived to compute the booms, and ended up with this vector of offsets to add to the shot 10 timestamps:
0.1450 0.0000 0.1902 0.0073 0.2176 0.0013 0.0001
So, my question to you @vt1 is how are you handling the timestamps for shot 10? What crack-boom corrections are you applying?
There are two other observations that make it implausible that shot 10 came from the North Barn area.
- If you listen to the NTD and Podium audio, it’s fairly easy to compare shots 9 and 10 and conclude that we hear a “boom” for shot 9 but a supersonic “crack” for shot 10. So, NTD and Podium were behind the line of fire for shot 9 but ahead of it for shot 10.
- The area by the North Barn entrance door has an obstructed line-of-sight to the shooter’s position on AGR6. It’s blocked by the thickest part of the tree, as seen in this drone shot:
Also, since I’m using the boom time for TMX, this gives me a lot of confidence in the TMX/Podium hyperbola. If I use my location for TMX shot 10 (586724.39, 4523506.91), I get nice alignment on the South Barn:
If I use your location for TMX shot 10 (586718.39, 4523512.20), the TMX hyperbola moves Northeast but still does not get very close to the North Barn. So, it doesn’t seem like your result for shot 10 should have turned out the way it did.
