Peter said “I don’t think EASA will ever allow PMA because it would do its Part 21 mates out of work. But also you can fit FAA-PMA parts to EASA-regs, anyway (despite claims to the contrary, it goes on all the time, openly, though perhaps not on AOC planes). Mine had loads of PMA parts put in by a UK Level 2 (?) 145 company, when it was G-reg”
You can legitimately use FAA PMA parts on any EASA aircraft so long as the aircraft type is specifically listed on the FAA PMA approval certificate for the part, and the 8130-3 states that the part is ‘Non- critical’. It’s addressed in the EU/US bi-lateral and EASA have published some specific guidance about it.
Peter wrote:
Autopilots need a lot of flight testing
Think about how much fun could be had “having” to fly many nice planes for free 
Peter wrote:
Autopilots need a lot of flight testing
And this is a big part of experimental and microlight aviation. People are willing to test new stuff. I don’t see why autopilots need a lot of flight testing though, unless you are thinking super advanced stuff.
And even the advanced stuff does not neccesarily be tested in flight. I remember the ‘engineering simulator’ at Fokker doing thousands of hours of autopilot flying with no-one there.
An autopilot needs to be tested for stability margins, all around the loading envelope and at various speeds between Vs and Vne.
It’s not a one day job.
Especially in planes with marginal stability i.e. many homebuilts, which don’t need to meet the monotonic (let alone progressive) control force criteria, or Vs doesn’t really exist because of insufficient control surface authority at max torque, so you have to pick some higher figure as Vs and placard the autopilot accordingly. Vne testing must be even more fun – I would imagine it is flutter limited in many cases.
I wonder if anyone has done self learning control loops? They are fairly common in industrial temperature controllers now, although in those applications there is plenty of time to make it work.
Peter wrote:
Vne testing must be even more fun – I would imagine it is flutter limited in many cases.
In the UK a Vne-dive is part of the annual PTF-renewal for the whole LAA-fleet.
While the experimental/homebuilt/ultra-light market across the USA and Europe is quite large and could always do with more manufacturers offering better products, I would say the trend of late has been away from standalone instruments and towards integrated systems such as offered by Garmin, Dynon and MGL.
This is certainly true of the RV scene, where the Garmin G3X and Dynon Skyview systems seem to be the most popular fits. These are sophisticated systems which have integrated GPS, autopilots (with envelope protection), engine management and fuel flow options.
There are also some pretty good standalone digital autopilots available, particularly those offered by Trutrak, which until Garmin introduced its own integrated system for the G3X was the autopilot of choice for the majority of the RV fleet.
I think autopilots would not be a great field to get into because the competition has a lot of experience under it’s belt, which is always a problem.
One would have to look for simpler standalone things.
Sure; glass products have taken a lot of “meat” out of the avionics business. So, which areas are poorly addressed? You want something which everybody should have but which nobody is doing well. That’s what I always did in business.
Incidentally, Trutrak may be a major player but this might concern some wishing to get their UK LAA plane IFR approved:
Peter wrote:
So, which areas are poorly addressed? You want something which everybody should have but which nobody is doing well.
Electronic fuses. It’s not all that well known a product, and only two are producing it. It is something that easily is integrated with glass also, and complement it perfectly. But, it is very well addressed by the two who makes them (vertical powers and MGL). They are in the US and SA, no one in Europe (as far as I know).
That and VOR/ADF/DME/TACAN in a nice compact packaging, maybe a nice digital display (connected through WiFi to a tablet/phone maybe and/or to the EFIS)
You cannot make anything that radiates anything (of aviation relevance) that is not approved.
Whether it needs a TSO is another matter but likely of no practical consequence. I think that for business purposes the bottom fell out of the navigation market about 15 years ago, due to GPS.
So that rules out DME TXP TACAN (well, the DME portion of a TACAN; the other half is a UHF VOR).
And a homebuilder can buy a DME on US Ebay, probably for similar money to a hypothetical non-approved product.
VOR/ADF is easy to do, though the ADF antenna is always going to be a big lump. Would someone really pay for it? There is no regulatory requirement on a homebuilt, though there would be if it is to get IFR approval.
So this may be an interesting area to look at: IFR
In terms of enforcement there is no driver for VOR or ADF. Cirruses have been flying in fair numbers around Europe for 10+ years with no DME and no ADF, flying NDB and “mandatory DME” approaches, with no known busts. And they hardly keep a low profile… homebuilt IFR does keep a very low profile – as is evident from any FR24 monitoring of the usual types. The Eurocontrol IFR profile of RVs and Lancairs is very close to zero. And the in-CAS profile of RVs and Lancairs at resasonably high altitudes (where a TXP would be mandatory) is very close to zero, too, once away from the Alps where there is quite a lot of it.
There is a really small “nav backup” point in a VOR receiver, but really small.
What would definitely be good is ILS, for obvious reasons. But surely there are “COM/NAV” radios which do all this? OTOH developing a VHF radio is a lot of work.
