The only reliable parachute ejection I have had so far in my various tetrahedra is sideways... what I have named "Lateral Thinking Technology (patent pending)" in a vehicle called Braques Big Black Breast.
You can see it at
https://www.palimpsest.co.uk/tet/tet.html
I had two singularly unsuccessful early flights with the big black one with rear exit... mostly owing to the chute being too tight which was due to the tube being rather too short for the chute ;-)
https://www.palimpsest.co.uk/KLOB/KLOB.html
I recently tried another rear exit design called Nothing2Hide last weekend but the base drag literally sucked the chute out which was poetic as I thought the only gremlin in those earlier flights was packing it too tight ;-)
Good thing it was undersized. N2H went on to a abbreviated apogee trailing its chute and returned to earth scarcely marred. A wet wipe and 3 inches of tape fixed it to fly again (once I have redone the chute deployment).
I have had two spacks using integral ejection charges (can't get the delay short enough with either Pro38 or Aerotech) but three perfectly timed ejections using a magnetic apogee detector and Pratt ejection canisters have made me change my ways.
I would have great reservations about the truncated tet within the big tet nosecone tho the idea has some appeal esp. since it would seem to allow a generous compartment for the chute.
An early ejection would be fighting against the huge base suction generated by the geometry's aerodynamics. A late ejection would require a charge big enough to eject the truncated inner tet >upwards< away from the plummeting outer tet.
I think that forward ejection is the way to go - I will be able to give you some empirical results as soon as I have refitted N2H.
I think you had a problem b/c of the tape. Tape, even lo tack masking tape, in tension is surprisingly strong (I learned that the hard way).
I think that a little attention to the chute exit door shape will solve your problem. It needs to be shaped so that it is held in position by the airstream on the way up yet free to fly off with some BP encouragement at apogee or afterwards.
The nice thing about tets (or pyramids if you insist) is that they have the terminal velocity of a hay bale. My first tet flight on a G spacked but took only 15 minutes with some 5 min epoxy and a file to prep for a second try with an H... which also spacked b/c the delay was too long and the chute was still too tight.
Short delays (and/or MAD triggers), loosely packed chutes, and any egress except rearwards are the ingredients for success IMHO
YMMV
Paul
Captain Tetrahedron
Team Uranus
UKRA 1276 Level 2