I don’t think that will be necessary. How each nozzle performs in thrust and IR suppression will be pretty evident from the engine tests alone. What will determine the TVC type, if it hasn’t already been decided, will be based on what requirements the PLAAF defines. The FCS integration for a 2D and 3D system will be very different. Trying to do a fly off of the two is a lot of unnecessary work to observe performance that can already be determined from static test beds.
You could simulate 2D TVC by locking in only single axis vectoring, but the control laws would still have to be very different. TVC FCS integration won’t just involve laws for when to deflect the nozzles, but also modified laws for the deflection of the control surfaces when nozzles are deflected as well.Furthermore, if they have a full 3D TVC engine installed, they could always software limit the FCS to simulate 2D TVC and potentially do the full flight test of how the J20 might perform with 2D WS15s.
The only thing they may want/need full sized live testing might be RCS testing, but that could easily be done on the ground with a repurposed prototype.
A fly off probably isn’t strictly necessary, but it would be cool to have, so forgive my wishful thinking here.
Just wanted to be very clear about just how much work it would be to develop both a 2D TVC FCS and a 3D TVC FCS for a fly off. My main point is just that it makes a lot more sense to figure out acceptable thrust loss and IR suppression requirements, pick one, and do the whole development that way rather than try to do both and see how they perform IRL. My thinking is that 3D TVC has already been chosen because of minimal thrust loss (based on the paper about the TVC used in the J-10 thrust lost might be as low as 3-5%?) and because there are probably other ways of getting comparable IR suppression without a deeply nested 2D design based on ambient air mixing methods.
