Peter wrote:
But it’s an interesting mental exercise
It’s a “metal” exercise to be sure 
Over the years I have seen countless FAA docs listing which installations need an STC, but whether they have absolute regulatory force I don’t know.
I don’t see flash rivet replacement as easy since you need to dimple the upper skin and countersink the lower skin, which means taking apart the entire aircraft. But it’s an interesting mental exercise – the sort of thing I like
If I wanted a Lancair IV I would have one already. All these homebuilts are useless to me in Europe because they can’t do legal IFR in most places. The TTX is good but it has the wrong sort of landing gear 
Anyway I don’t want to digress.
But back to the OP:
If you wanna go fast & efficient, sell the TB20 and get yerself a Lancair/Columbia/Ttx
… But all that is just theory. The main difference between the STC process and the Field Approval is the STC is reviewed by the Aircraft Certification Office (ACO) rather than the FSDO. But, in reality, if the FSDO considers your Field Approval modification a bit too complex for them to vet, they most likely will kick it over to the ACO and now were talking 1-time STC.
Clear as mud, right ?
… The point being is that if you are going to develop something as sophisticated as an autopilot system or EFIS then you might as well get the STC rather than have the installers have to sweat a Field Approval.
To my knowledge, there is nothing in the FARs that state which class of equipment MUST be installed under STC versus Field Approval.
Peter wrote:
So you can install an autopilot under a field approval?
Sure, if you can get one !
So you can install an autopilot under a field approval?
Peter wrote:
Certain mods require an STC anyway. EFIS, autopilot, what else?
Only because it would be next to impossible to sell anything like those SANS a proper STC, and better yet, an AML STC !
BTW: A couple of things to think about :
- The wing profile on the TB Series is NOT considered a Laminar Profile, so cleaning up the surface will have little to no affect on speed efficiency;
- Flush rivets require that the supporting sheets be counter-sunk by either drilling or dimpling. Both techniques require different sheet thickness to avoid cracking and/or deformations. A good example of this is found on the Cessna 210 and 177 wings. Originally developed for the heavier 210 when the engineers had to use thinner wing skins for the lighter 177 they found that the dimpling process was introducing alot of deformations and waviness which virtually destroyed the benefit od using flush rivets !
/
