The Elusive LPR Whistle Rocket, Part 2

My Odd'l Rocket Cyclone model does whistle on the way up. It's not loud, but can be heard during the coast phase of the engine burn.
The Cyclone has a hole punched in the side of the body tube to allow some of the ejection charge to vent like the old Estes Scout. That hole along with the the hollow area above the engine is the right area to make an audible whistle. The air flow is split at the low edge of the punched hole.
I tried to take this idea of a split air stream to make a louder whistle. The whistle has to be loud enough to be heard from the ground.

I thought maybe strategically placed hole might get a good whistle, but I didn't want to go through a dozen test flight variations.
Then it hit me, I could get a whistle while dong a swing test.


Here's my test model, a simple 3FNC rocket. My plan was to have three of these around a central core. The central core would have the engine and parachute. Each of the three "strap-on" whistlers could be tuned to a different note possibly making up a major chord triad of Do-Mi-So.

A Kevlar swing string was tied an taped on at the C/G point.




In the back is a capped engine block. The block could be slid up and down for "tuning". The capped block wasn't glued in. Using a dowel I could change the interior cavity size by sliding it up and down inside the tube. The right position might find a whistle "sweet spot".

 

Here's some of the tests - 

Test 1: A small hole was drilled near the top of the nose cone. The base of the nose cone was cut off to allow the air to past through to the body tube. The model was swing tested at a fast enough speed where a whistle should happen. Very little wind noise was heard.

Test 2: This upper nose cone hole was taped over and a hole was punched in the body tube. Swing testing again led to wind noise and deflection but no real whistle.

Test 3: The top of the nose cone was cut off at an angle. More swing tests got a whistle, but not loud enough.

Test 4: This cut make the most noise but still not loud enough to be heard over the engine.
The problem is the speed of the rocket has to be consistent to activate the perfect whistle design. Who knows what speed and whistle shape could work together to produce a audible result?

I had hope for some better results this time. I'm only out an hour of time and three nose cones.