Message posted by RCS polemodel on May 11, 2009 at 10:03:02 PST:
I don't know how else to say it: this IS the cheaper solution they engineered. The 757 carries the full set of F-22 electronic hardware and software--the radar, the computers, the sensors which you see stuck on the outside of the 757, the data link, etc. And it carries a full F-22 cockpit, in which sits a pilot who operates all that stuff. And it carries a crew of RF and software engineers who can monitor and troubleshoot the behavior of the systems in real time. (Well, not always real time: since the F-22 systems aren't responsible for keeping themselves in the air, the software guys can run things in 'extreme slow motion' and watch each step in the software execution with their own eyeballs.) That is, they can find code errors and correct them literally on the spot, while the "F-22" inside the 757 is still in the air, then re-run the exercise to confirm that the rewritten code works.
The alternative is to fit, as you said, a data acquisition package into the weapons bay of a production F-22 to record what the systems do as the pilot executes a test protocol. Then he lands, the data are unloaded, and the software people plow through it to see if anything went wrong and if so, where the code errors are. Then they re-write some code, re-load the software on the F-22, and sent it up again. This would have added years and millions to the program, even if, as was considered, you divert a bunch of F-22s from the production pipeline to devote to this effort. In which case you've also got the additional nightmare of managing multiple versions of the software concurrently.
The flying testbed approach is the right one. Lockheed and Northrop have copied it with the F-35, and Boeing plans to make a fortune selling time on the 757 to other programs when the F-22 work is complete.
No argument, modern combat aircraft are ridiculously expensive, and the development programs are a huge part of that. I have long held that a good way to make the process more efficient would be to get more quality work out of RF engineers by tripling their pay and sending them on long vacations to exotic locales. But I'm not an accountant ;-)
In Reply to: Re: Strange plane at Nellis Red Flag posted by bubba on May 11, 2009 at 9:24:38 PST:
Replies: