AeroFranz said:
VG buys its way onto the aircraft anytime you have to live in EXTREMELY different corners of the flight envelope.
Fixed geometry simply will not allow you to do high-speed low altitude penetration AND economic high-altitude loiter, just to pick examples. In the conceptual design phase you analyse typical mission profiles and trade weight and complexity. If the requirements are really that far apart, then VG is not necessarily heavier; hell, it may be the only solution.
The fact that no manufacturer has bothered to do more than two VG airplanes in a row is more a reflection of changing mission requirements than "fashion" or technology.
This is pretty much what I would have said had I got back here in time, so now, if you'll permit me, I'll philosophize in my usual unnecessarily wordy style regarding Radical's question.
VG does introduce complexity, but it can add a lot of capability. While a vg wing may not completely match the performance of a fixed wing at the fw's optimum point, it can offer very good performance over a larger speed because it has a larger "sweet spot". It can enhance loiter, as Sundog mentioned but it offers a lot more. like everything else in aviation, wing is a series of compromises necessary to be functional over a particular range. A wing optimized for high speed flight, such as an SR-71 is going to of necessity have poor low speed performance. Similarly, an A-10 wing is going to give you all kinds of trouble at the high subsonic, let alone supersonic, levels. With VG you can take a basic wing design, sweep it aft for higher speeds and then forward for good performance at lower speed performance without having to go to an incredibly complex design or high angle of attack for slowest speeds. That's one of the big reasons all the competitors except North American bidding for what would be my namesake (you forgot that one, Jemiba!) went for vg. It's also why you see the B-1, TU-22M and TU-160 use vg. It allows them to optimize for a high speed low drag platform and yet still have acceptable airfield performance.
In the case of the Tomcat, it also was used to give it good maneuverability over a larger speed range., Although it can't match the Hornet's high AoA performance at low speeds (although until the F-22 the Hornet was the only fighter that had better high AoA performance), it does well slow and as speed picks up it re-optimizes the wing planform for better performance over a larger range. And, the F-14 could come aboard the boat slower. That re-optimizing helps throughout the envelope. That's why the primary mode of wing sweep is full auto (based on speed, power and what the a/c is doing at the moment), a characteristic only shared by the Tornado F3, and even there it doesn't do it as well. The vg also allows it to become essentially a pure delta, with the consequent dramtic lowering of wing loading as the lifting body tunnel comes into play. Frankly, to build a fixed wing in the '70s and '80s that had as much versatility as the Tomcat's would have been heavier than a vg. The downside of the full auto wing is that close in, its position and movement telegraphs what the Tomcat is about to do to those who have studied the a/c. I think that's what you meant, Sundog, about giving away your energy state. I just prefer to express it this way because the wing can move for reasons other than just energy state. Again, I'm referring to the F-14 because it's the one that uses wing position as a maneuvering device.
It seems that nowadays we aren't asking to optimize over such a large range. VG also is harder "to stealth". The B-2 lumbers compared to the B-1, but agility was not a driver for it relative to the former. It can still be done, it's just harder. For example, in Lockheed's vg design for the late, lamented A/FX program they were willing to trade of some of the agility of the Navy version for more stealth in the Air Force version.
Nowadays, whether through more advanced technology, changing design philosophies (carry a ton of fuel, have gobs of power or in the case of USAF assume you'll always have 8,000 ft. of runway and a golf course
) or different requirements (don't need to optimize over an entire speed range) it appears that for now designers don't feel vg is worth the complexity.