Some parts of retractable gear cost over $20,000 today and yet Cessna has difficulties producing them
What is the actual "difficulty"?
I bet it is somebody just being stupid.
Probably, virtually nobody is buying them, and the accountants decided to not stock it, and they never got around to sorting it out.
Same for the $50k Malibu part.
There is nothing on any GA plane which is even remotely difficult to manufacture.
They said that they determined there would be no market for the retractable due to very high cost they would have to sell at.
But they would say that, because they don't have any retractable models which would sell anyway (for other reasons).
If you have some insight into how the Cessna spare part business operates, it becomes more plausible.
Cessna et al do very little themselves, never have. Cutting sheet aluminium and riveting, that's it. For the rest, they have suppliers. Often, they just don't find a company that has the required certifications and is willing to produce the part. Then there is the liability thing. A lot of suppliers to GA aircraft makers have deliberately given up in fear of liability.
The $50k Malibu part was because only one company was willing to produce it. Then with a lot of effort, Piper managed to convince a second company and suddenly the price dropped. A financially weak company (read: all GA aircraft makers) cannot afford to make a big order and keep stock for the demand of the next 30 years. And even that 30 years of supply would still be a very small order. Whenever Cessna runs out of inventory, you typically see a steep price increase. Their prices do reflect the 3rd party cost, Cessna is very much trying to keep the prices down and thus the fleet viable and in turn the spare part business. In the 1986-1996 time, Cessna turned a profit in their GA division, only because of the large fleet (ca 140,000) and I am sure that it is the same today, the money does not come from producing new GA aircraft but servicing the fleet.
Yes; it is not that different from Socata though they have a lot of CNC facilities in-house.
But I still say the problem is caused by inept organisation coupled with low volumes for many items.
If you are a Part 145 business, plus in-house-DER (EASA 21 in Euro speak) then you have no need to buy "aviation certified" (whatever that means) parts from outside.
You can buy industrial-grade parts, with batch traceability paperwork, and certify them yourself.
One real problem is that the industrial part makers don't want to sell small batches, with the paperwork. For example an alternator is just the same as a car/truck one, and you have to twist the arm of the alternator maker to sell you just 100 of them, with special paperwork. So a $50 (trade price) alternator gets sold to the plane maker for say $500.
Actually an alternator is a lousy example because nobody with a brain is going to buy an alternator from the plane maker! You buy one of the numerous PMA parts. Same as nobody does (or should!!) buy normal service items from the plane maker: spark plugs, filters, lubes, seals, engine accessories incl. vac pumps, belts, hoses, etc.
But hey the plane maker can buy the PMA parts too 
But there aren't many such parts that add up to the huge price tag of something like the Cessna TT. On that, or a TB20 for that matter, or an SR22, most of the cost is labour in building the airframe by hand, and the huge cost of bought-in major items: engine and avionics.
I often hear stories about such and such being a huge cost (product liability insurance is a favourite one) but it is never supported by real data. I recall somebody digging around Garmin published accounts for their insurance costs, and they were in there, and they were miniscule (as I would expect).
The final nail in the coffin is that GA was huge in the 1960s and 1970s, with volumes 10x bigger than in later decades, yet most of the players are still working as if nothing changed. The management structures at say Honeywell (Bendix/King) are still in the 1960s. I bet they have a 1000-seat corporate license for Powerpoint 
I thought the reason behind the fixed gear on the Cirrus was to make a crash on the parachute survivable
You can buy industrial-grade parts, with batch traceability paperwork, and certify them yourself.
So you ask a random Chinese to cast a vital gear part and you just certify and sell that as Cessna? You are greatly underestimating the risks involved and the diligence required. Remember the Lycoming crankshafts?
A friend of mine built a very fancy architect house. It took him 6 months to find a glass manufacturer that was willing to take the job. The company charged an outrageous price and yet they deeply regretted taking on the job. The project was full of complications and they ended up losing money. I can understand why Cessna etc. have trouble finding suppliers. Often these companies have to be very competent and prefer to not get involved in such a low volume business, no matter what it pays.
The alternator/pump/etc. example also does not apply because those parts are not airframe dependent, there is a much bigger market. The Lycoming/Continental spare part / accessory market is still a working market with good predictability.
I thought the reason behind the fixed gear on the Cirrus was to make a crash on the parachute survivable
You don't need a fixed gear for that. Connect the gear extension switch (most light types have electrical or electro-hydraulic gear mechanisms) to the chute deployment system and it will automatically come out when needed. No rocket science involved.
And regarding the fixed/retractable gear discussion: As already stated earlier this week in another thread I do not think that a retractable gear is worth the extra cost and complication in anything that does not exceed 200KT. An aerodynamically well-designed gear does not need to have a lot of drag. Just do not compare this to the drag of a retractable gear that is not designed with aerodynamics is mind at all!
So you ask a random Chinese to cast a vital gear part and you just certify and sell that as Cessna? You are greatly underestimating the risks involved and the diligence required.
No, you use a random American company
It will depend on what it is.
For example Socata (and many other French suppliers of Airbus parts) make a lot of stuff down in Morocco. QA down there must be, ahem, interesting... but these won't be complex parts. They will be bits of bent-up ally, riveted into subassemblies, which are very labour intensive to make in the First World. Most of a GA plane is like that - simple but time consuming stuff. Very easy to inspect back at the factory.
Remember the Lycoming crankshafts?
Yeah - $14k
But that was caused by (a) the subcontractor not doing some heat treatment stages, due to stupidity, despite being an American company (which led to some in-flight breakages) and (b) Lyco authorising the subcontractor to use vanadium (which led to the 12 year life limit AD).
Anybody could do that. You cannot safeguard against what is basically fraud (document forgery). Aviation is full of that, from screws all the way up to pilot licenses.
As already stated earlier this week in another thread I do not think that a retractable gear is worth the extra cost and complication in anything that does not exceed 200KT
I think that will always be debatable, and really hangs on one's view of whether another ~15kt is worth having.
Many would say it isn't. Obviously Cirrus will say it isn't - in fact they tend to say it's only 3kt
Personally, I think 15kt is worth having.
The issue goes up exponentially as one goes faster, which is why reaching 200kt is so very very hard with a piston engine - unless you either pour in fuel (a Spitfire e.g.), make the cockpit so intimate you are going to make a female passenger pregnant, or fly very high.
Nothing in light GA gets anywhere remotely near 200kt at normal GA altitudes, so for that you are probably right.
Nothing in light GA gets anywhere remotely near 200kt at normal GA altitudes, so for that you are probably right.
Panthera If this project works out, then it's the proof that aerodynamics (and that includes not dragging a gear) do matter. So far it's been pretty disappointing. Given a certain fuel flow and useful load, a "modern" airplane like a Cirrus or TB20 do not or only marginally outperform a 1950s airframe. And a 1997 C172 is a lame duck compared to a 1976 C172 despite having more horsepower.
My memory isn't good but are they really claiming 200kt+ IAS?
No but why is IAS relevant? It's not the speed of the aircraft and it's something from the stone age when there was no other way way to determine airspeed.
I suggest you get yourself an Air Data Computer (ADC), then you will quickly realize that IAS is completely meaningless
